if(BUILD_ZLIB)
ocv_clear_vars(ZLIB_FOUND)
else()
+ ocv_clear_internal_cache_vars(ZLIB_LIBRARY ZLIB_INCLUDE_DIR)
find_package(ZLIB "${MIN_VER_ZLIB}")
if(ZLIB_FOUND AND ANDROID)
if(ZLIB_LIBRARIES MATCHES "/usr/(lib|lib32|lib64)/libz.so$")
if(BUILD_JPEG)
ocv_clear_vars(JPEG_FOUND)
else()
+ ocv_clear_internal_cache_vars(JPEG_LIBRARY JPEG_INCLUDE_DIR)
include(FindJPEG)
endif()
if(NOT JPEG_FOUND)
- ocv_clear_vars(JPEG_LIBRARY JPEG_LIBRARIES JPEG_INCLUDE_DIR)
+ ocv_clear_vars(JPEG_LIBRARY JPEG_INCLUDE_DIR)
if(NOT BUILD_JPEG_TURBO_DISABLE)
set(JPEG_LIBRARY libjpeg-turbo CACHE INTERNAL "")
if(BUILD_TIFF)
ocv_clear_vars(TIFF_FOUND)
else()
+ ocv_clear_internal_cache_vars(TIFF_LIBRARY TIFF_INCLUDE_DIR)
include(FindTIFF)
if(TIFF_FOUND)
ocv_parse_header("${TIFF_INCLUDE_DIR}/tiff.h" TIFF_VERSION_LINES TIFF_VERSION_CLASSIC TIFF_VERSION_BIG TIFF_VERSION TIFF_BIGTIFF_VERSION)
if(BUILD_WEBP)
ocv_clear_vars(WEBP_FOUND WEBP_LIBRARY WEBP_LIBRARIES WEBP_INCLUDE_DIR)
else()
+ ocv_clear_internal_cache_vars(WEBP_LIBRARY WEBP_INCLUDE_DIR)
include(cmake/OpenCVFindWebP.cmake)
if(WEBP_FOUND)
set(HAVE_WEBP 1)
if(BUILD_PNG)
ocv_clear_vars(PNG_FOUND)
else()
+ ocv_clear_internal_cache_vars(PNG_LIBRARY PNG_INCLUDE_DIR)
include(FindPNG)
if(PNG_FOUND)
include(CheckIncludeFile)
if(WITH_OPENEXR)
ocv_clear_vars(HAVE_OPENEXR)
if(NOT BUILD_OPENEXR)
+ ocv_clear_internal_cache_vars(OPENEXR_INCLUDE_PATHS OPENEXR_LIBRARIES OPENEXR_ILMIMF_LIBRARY OPENEXR_VERSION)
include("${OpenCV_SOURCE_DIR}/cmake/OpenCVFindOpenEXR.cmake")
endif()
endforeach()
unset(__depsvar)
- # hack for python
- set(__python_idx)
- list(FIND OPENCV_MODULE_${full_modname}_WRAPPERS "python" __python_idx)
- if (NOT __python_idx EQUAL -1)
- list(REMOVE_ITEM OPENCV_MODULE_${full_modname}_WRAPPERS "python")
- list(APPEND OPENCV_MODULE_${full_modname}_WRAPPERS "python_bindings_generator" "python2" "python3")
- endif()
- unset(__python_idx)
-
ocv_list_unique(OPENCV_MODULE_${full_modname}_REQ_DEPS)
ocv_list_unique(OPENCV_MODULE_${full_modname}_OPT_DEPS)
ocv_list_unique(OPENCV_MODULE_${full_modname}_PRIVATE_REQ_DEPS)
set(OPENCV_MODULES_DISABLED_USER ${OPENCV_MODULES_DISABLED_USER} "${the_module}" CACHE INTERNAL "List of OpenCV modules explicitly disabled by user")
endif()
- # add reverse wrapper dependencies
+ # add reverse wrapper dependencies (BINDINDS)
foreach (wrapper ${OPENCV_MODULE_${the_module}_WRAPPERS})
- ocv_add_dependencies(opencv_${wrapper} OPTIONAL ${the_module})
+ if(wrapper STREQUAL "python") # hack for python (BINDINDS)
+ ocv_add_dependencies(opencv_python2 OPTIONAL ${the_module})
+ ocv_add_dependencies(opencv_python3 OPTIONAL ${the_module})
+ else()
+ ocv_add_dependencies(opencv_${wrapper} OPTIONAL ${the_module})
+ endif()
+ if(DEFINED OPENCV_MODULE_opencv_${wrapper}_bindings_generator_CLASS)
+ ocv_add_dependencies(opencv_${wrapper}_bindings_generator OPTIONAL ${the_module})
+ endif()
endforeach()
# stop processing of current file
endforeach()
endmacro()
+
+# Clears passed variables with INTERNAL type from CMake cache
+macro(ocv_clear_internal_cache_vars)
+ foreach(_var ${ARGN})
+ get_property(_propertySet CACHE ${_var} PROPERTY TYPE SET)
+ if(_propertySet)
+ get_property(_type CACHE ${_var} PROPERTY TYPE)
+ if(_type STREQUAL "INTERNAL")
+ message("Cleaning INTERNAL cached variable: ${_var}")
+ unset(${_var} CACHE)
+ endif()
+ endif()
+ endforeach()
+ unset(_propertySet)
+ unset(_type)
+endmacro()
+
+
set(OCV_COMPILER_FAIL_REGEX
"argument .* is not valid" # GCC 9+ (including support of unicode quotes)
"command[- ]line option .* is valid for .* but not for C\\+\\+" # GNU
--- /dev/null
+set(OPENCV_SKIP_LINK_AS_NEEDED 1)
--- /dev/null
+getBlobFromImage = function(inputSize, mean, std, swapRB, image) {
+ let mat;
+ if (typeof(image) === 'string') {
+ mat = cv.imread(image);
+ } else {
+ mat = image;
+ }
+
+ let matC3 = new cv.Mat(mat.matSize[0], mat.matSize[1], cv.CV_8UC3);
+ cv.cvtColor(mat, matC3, cv.COLOR_RGBA2BGR);
+ let input = cv.blobFromImage(matC3, std, new cv.Size(inputSize[0], inputSize[1]),
+ new cv.Scalar(mean[0], mean[1], mean[2]), swapRB);
+
+ matC3.delete();
+ return input;
+}
+
+loadLables = async function(labelsUrl) {
+ let response = await fetch(labelsUrl);
+ let label = await response.text();
+ label = label.split('\n');
+ return label;
+}
+
+loadModel = async function(e) {
+ return new Promise((resolve) => {
+ let file = e.target.files[0];
+ let path = file.name;
+ let reader = new FileReader();
+ reader.readAsArrayBuffer(file);
+ reader.onload = function(ev) {
+ if (reader.readyState === 2) {
+ let buffer = reader.result;
+ let data = new Uint8Array(buffer);
+ cv.FS_createDataFile('/', path, data, true, false, false);
+ resolve(path);
+ }
+ }
+ });
+}
+
+getTopClasses = function(probs, labels, topK = 3) {
+ probs = Array.from(probs);
+ let indexes = probs.map((prob, index) => [prob, index]);
+ let sorted = indexes.sort((a, b) => {
+ if (a[0] === b[0]) {return 0;}
+ return a[0] < b[0] ? -1 : 1;
+ });
+ sorted.reverse();
+ let classes = [];
+ for (let i = 0; i < topK; ++i) {
+ let prob = sorted[i][0];
+ let index = sorted[i][1];
+ let c = {
+ label: labels[index],
+ prob: (prob * 100).toFixed(2)
+ }
+ classes.push(c);
+ }
+ return classes;
+}
+
+loadImageToCanvas = function(e, canvasId) {
+ let files = e.target.files;
+ let imgUrl = URL.createObjectURL(files[0]);
+ let canvas = document.getElementById(canvasId);
+ let ctx = canvas.getContext('2d');
+ let img = new Image();
+ img.crossOrigin = 'anonymous';
+ img.src = imgUrl;
+ img.onload = function() {
+ ctx.drawImage(img, 0, 0, canvas.width, canvas.height);
+ };
+}
+
+drawInfoTable = async function(jsonUrl, divId) {
+ let response = await fetch(jsonUrl);
+ let json = await response.json();
+
+ let appendix = document.getElementById(divId);
+ for (key of Object.keys(json)) {
+ let h3 = document.createElement('h3');
+ h3.textContent = key + " model";
+ appendix.appendChild(h3);
+
+ let table = document.createElement('table');
+ let head_tr = document.createElement('tr');
+ for (head of Object.keys(json[key][0])) {
+ let th = document.createElement('th');
+ th.textContent = head;
+ th.style.border = "1px solid black";
+ head_tr.appendChild(th);
+ }
+ table.appendChild(head_tr)
+
+ for (model of json[key]) {
+ let tr = document.createElement('tr');
+ for (params of Object.keys(model)) {
+ let td = document.createElement('td');
+ td.style.border = "1px solid black";
+ if (params !== "modelUrl" && params !== "configUrl" && params !== "labelsUrl") {
+ td.textContent = model[params];
+ tr.appendChild(td);
+ } else {
+ let a = document.createElement('a');
+ let link = document.createTextNode('link');
+ a.append(link);
+ a.href = model[params];
+ td.appendChild(a);
+ tr.appendChild(td);
+ }
+ }
+ table.appendChild(tr);
+ }
+ table.style.width = "800px";
+ table.style.borderCollapse = "collapse";
+ appendix.appendChild(table);
+ }
+}
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Image Classification Example</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Image Classification Example</h2>
+<p>
+ This tutorial shows you how to write an image classification example with OpenCV.js.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configFile</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Try it</b> button to see the result. You can choose any other images.<br>
+</p>
+
+<div class="control"><button id="tryIt" disabled>Try it</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <canvas id="canvasInput" width="400" height="400"></canvas>
+ </td>
+ <td>
+ <table style="visibility: hidden;" id="result">
+ <thead>
+ <tr>
+ <th scope="col">#</th>
+ <th scope="col" width=300>Label</th>
+ <th scope="col">Probability</th>
+ </tr>
+ </thead>
+ <tbody>
+ <tr>
+ <th scope="row">1</th>
+ <td id="label0" align="center"></td>
+ <td id="prob0" align="center"></td>
+ </tr>
+ <tr>
+ <th scope="row">2</th>
+ <td id="label1" align="center"></td>
+ <td id="prob1" align="center"></td>
+ </tr>
+ <tr>
+ <th scope="row">3</th>
+ <td id="label2" align="center"></td>
+ <td id="prob2" align="center"></td>
+ </tr>
+ </tbody>
+ </table>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ canvasInput <input type="file" id="fileInput" name="file" accept="image/*">
+ </div>
+ </td>
+ <td></td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="13" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.Main loop in which will read the image from canvas and do inference once.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Load labels from txt file and process it into an array.</p>
+ <textarea class="code" rows="7" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+ <p>6.The post-processing, including softmax if needed and get the top classes from the output vector.</p>
+ <textarea class="code" rows="35" cols="100" id="codeEditor5" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [224,224];
+mean = [104, 117, 123];
+std = 1;
+swapRB = false;
+
+// record if need softmax function for post-processing
+needSoftmax = false;
+
+// url for label file, can from local or Internet
+labelsUrl = "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt";
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+main = async function() {
+ const labels = await loadLables(labelsUrl);
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, 'canvasInput');
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const probs = softmax(result);
+ const classes = getTopClasses(probs, labels);
+
+ updateResult(classes, time);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+</script>
+
+<script id="codeSnippet5" type="text/code-snippet">
+softmax = function(result) {
+ let arr = result.data32F;
+ if (needSoftmax) {
+ const maxNum = Math.max(...arr);
+ const expSum = arr.map((num) => Math.exp(num - maxNum)).reduce((a, b) => a + b);
+ return arr.map((value, index) => {
+ return Math.exp(value - maxNum) / expSum;
+ });
+ } else {
+ return arr;
+ }
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_image_classification_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let loadLablesCode = 'loadLables = ' + loadLables.toString();
+ document.getElementById('codeEditor2').value = loadLablesCode;
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor3').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor4').value = loadModelCode;
+
+ utils.loadCode('codeSnippet5', 'codeEditor5');
+ let getTopClassesCode = 'getTopClasses = ' + getTopClasses.toString();
+ document.getElementById('codeEditor5').value += '\n' + '\n' + getTopClassesCode;
+
+ let canvas = document.getElementById('canvasInput');
+ let ctx = canvas.getContext('2d');
+ let img = new Image();
+ img.crossOrigin = 'anonymous';
+ img.src = 'space_shuttle.jpg';
+ img.onload = function() {
+ ctx.drawImage(img, 0, 0, canvas.width, canvas.height);
+ };
+
+ let tryIt = document.getElementById('tryIt');
+ tryIt.addEventListener('click', () => {
+ initStatus();
+ document.getElementById('status').innerHTML = 'Running function main()...';
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ if (modelPath === "") {
+ document.getElementById('status').innerHTML = 'Runing failed.';
+ utils.printError('Please upload model file by clicking the button first.');
+ } else {
+ setTimeout(main, 1);
+ }
+ });
+
+ let fileInput = document.getElementById('fileInput');
+ fileInput.addEventListener('change', (e) => {
+ initStatus();
+ loadImageToCanvas(e, 'canvasInput');
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ tryIt.removeAttribute('disabled');
+ });
+
+ var main = async function() {};
+ var softmax = function(result){};
+ var getTopClasses = function(mat, labels, topK = 3){};
+
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+ utils.executeCode('codeEditor5');
+
+ function updateResult(classes, time) {
+ try{
+ classes.forEach((c,i) => {
+ let labelElement = document.getElementById('label'+i);
+ let probElement = document.getElementById('prob'+i);
+ labelElement.innerHTML = c.label;
+ probElement.innerHTML = c.prob + '%';
+ });
+ let result = document.getElementById('result');
+ result.style.visibility = 'visible';
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('result').style.visibility = 'hidden';
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+{
+ "caffe": [
+ {
+ "model": "alexnet",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "needSoftmax": "false",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt",
+ "modelUrl": "http://dl.caffe.berkeleyvision.org/bvlc_alexnet.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/BVLC/caffe/master/models/bvlc_alexnet/deploy.prototxt"
+ },
+ {
+ "model": "densenet",
+ "mean": "127.5, 127.5, 127.5",
+ "std": "0.007843",
+ "swapRB": "false",
+ "needSoftmax": "true",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt",
+ "modelUrl": "https://drive.google.com/open?id=0B7ubpZO7HnlCcHlfNmJkU2VPelE",
+ "configUrl": "https://raw.githubusercontent.com/shicai/DenseNet-Caffe/master/DenseNet_121.prototxt"
+ },
+ {
+ "model": "googlenet",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "needSoftmax": "false",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt",
+ "modelUrl": "http://dl.caffe.berkeleyvision.org/bvlc_googlenet.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/BVLC/caffe/master/models/bvlc_googlenet/deploy.prototxt"
+ },
+ {
+ "model": "squeezenet",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "needSoftmax": "false",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt",
+ "modelUrl": "https://raw.githubusercontent.com/forresti/SqueezeNet/master/SqueezeNet_v1.0/squeezenet_v1.0.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/forresti/SqueezeNet/master/SqueezeNet_v1.0/deploy.prototxt"
+ },
+ {
+ "model": "VGG",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "needSoftmax": "false",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt",
+ "modelUrl": "http://www.robots.ox.ac.uk/~vgg/software/very_deep/caffe/VGG_ILSVRC_19_layers.caffemodel",
+ "configUrl": "https://gist.githubusercontent.com/ksimonyan/3785162f95cd2d5fee77/raw/f02f8769e64494bcd3d7e97d5d747ac275825721/VGG_ILSVRC_19_layers_deploy.prototxt"
+ }
+ ],
+ "tensorflow": [
+ {
+ "model": "inception",
+ "mean": "123, 117, 104",
+ "std": "1",
+ "swapRB": "true",
+ "needSoftmax": "false",
+ "labelsUrl": "https://raw.githubusercontent.com/petewarden/tf_ios_makefile_example/master/data/imagenet_comp_graph_label_strings.txt",
+ "modelUrl": "https://raw.githubusercontent.com/petewarden/tf_ios_makefile_example/master/data/tensorflow_inception_graph.pb"
+ }
+ ]
+}
\ No newline at end of file
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Image Classification Example with Camera</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Image Classification Example with Camera</h2>
+<p>
+ This tutorial shows you how to write an image classification example with camera.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configFile</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Start/Stop</b> button to start or stop the camera capture.<br>
+</p>
+
+<div class="control"><button id="startAndStop" disabled>Start</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <video id="videoInput" width="400" height="400"></video>
+ </td>
+ <td>
+ <table style="visibility: hidden;" id="result">
+ <thead>
+ <tr>
+ <th scope="col">#</th>
+ <th scope="col" width=300>Label</th>
+ <th scope="col">Probability</th>
+ </tr>
+ </thead>
+ <tbody>
+ <tr>
+ <th scope="row">1</th>
+ <td id="label0" align="center"></td>
+ <td id="prob0" align="center"></td>
+ </tr>
+ <tr>
+ <th scope="row">2</th>
+ <td id="label1" align="center"></td>
+ <td id="prob1" align="center"></td>
+ </tr>
+ <tr>
+ <th scope="row">3</th>
+ <td id="label2" align="center"></td>
+ <td id="prob2" align="center"></td>
+ </tr>
+ </tbody>
+ </table>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ videoInput
+ </div>
+ </td>
+ <td></td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="13" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.The function to capture video from camera, and the main loop in which will do inference once.</p>
+ <textarea class="code" rows="35" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Load labels from txt file and process it into an array.</p>
+ <textarea class="code" rows="7" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+ <p>6.The post-processing, including softmax if needed and get the top classes from the output vector.</p>
+ <textarea class="code" rows="35" cols="100" id="codeEditor5" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [224,224];
+mean = [104, 117, 123];
+std = 1;
+swapRB = false;
+
+// record if need softmax function for post-processing
+needSoftmax = false;
+
+// url for label file, can from local or Internet
+labelsUrl = "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/classification_classes_ILSVRC2012.txt";
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+let frame = new cv.Mat(video.height, video.width, cv.CV_8UC4);
+let cap = new cv.VideoCapture(video);
+
+main = async function(frame) {
+ const labels = await loadLables(labelsUrl);
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, frame);
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const probs = softmax(result);
+ const classes = getTopClasses(probs, labels);
+
+ updateResult(classes, time);
+ setTimeout(processVideo, 0);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+
+function processVideo() {
+ try {
+ if (!streaming) {
+ return;
+ }
+ cap.read(frame);
+ main(frame);
+ } catch (err) {
+ utils.printError(err);
+ }
+}
+
+setTimeout(processVideo, 0);
+</script>
+
+<script id="codeSnippet5" type="text/code-snippet">
+softmax = function(result) {
+ let arr = result.data32F;
+ if (needSoftmax) {
+ const maxNum = Math.max(...arr);
+ const expSum = arr.map((num) => Math.exp(num - maxNum)).reduce((a, b) => a + b);
+ return arr.map((value, index) => {
+ return Math.exp(value - maxNum) / expSum;
+ });
+ } else {
+ return arr;
+ }
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_image_classification_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let loadLablesCode = 'loadLables = ' + loadLables.toString();
+ document.getElementById('codeEditor2').value = loadLablesCode;
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor3').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor4').value = loadModelCode;
+
+ utils.loadCode('codeSnippet5', 'codeEditor5');
+ let getTopClassesCode = 'getTopClasses = ' + getTopClasses.toString();
+ document.getElementById('codeEditor5').value += '\n' + '\n' + getTopClassesCode;
+
+ let video = document.getElementById('videoInput');
+ let streaming = false;
+ let startAndStop = document.getElementById('startAndStop');
+ startAndStop.addEventListener('click', () => {
+ if (!streaming) {
+ utils.clearError();
+ utils.startCamera('qvga', onVideoStarted, 'videoInput');
+ } else {
+ utils.stopCamera();
+ onVideoStopped();
+ }
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ startAndStop.removeAttribute('disabled');
+
+ });
+
+ var main = async function(frame) {};
+ var softmax = function(result){};
+ var getTopClasses = function(mat, labels, topK = 3){};
+
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+ utils.executeCode('codeEditor5');
+
+ function onVideoStarted() {
+ streaming = true;
+ startAndStop.innerText = 'Stop';
+ videoInput.width = videoInput.videoWidth;
+ videoInput.height = videoInput.videoHeight;
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ }
+
+ function onVideoStopped() {
+ streaming = false;
+ startAndStop.innerText = 'Start';
+ initStatus();
+ }
+
+ function updateResult(classes, time) {
+ try{
+ classes.forEach((c,i) => {
+ let labelElement = document.getElementById('label'+i);
+ let probElement = document.getElementById('prob'+i);
+ labelElement.innerHTML = c.label;
+ probElement.innerHTML = c.prob + '%';
+ });
+ let result = document.getElementById('result');
+ result.style.visibility = 'visible';
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('result').style.visibility = 'hidden';
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Object Detection Example</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Object Detection Example</h2>
+<p>
+ This tutorial shows you how to write an object detection example with OpenCV.js.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configFile</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Try it</b> button to see the result. You can choose any other images.<br>
+</p>
+
+<div class="control"><button id="tryIt" disabled>Try it</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <canvas id="canvasInput" width="400" height="400"></canvas>
+ </td>
+ <td>
+ <canvas id="canvasOutput" style="visibility: hidden;" width="400" height="400"></canvas>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ canvasInput <input type="file" id="fileInput" name="file" accept="image/*">
+ </div>
+ </td>
+ <td>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile" name="file">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="15" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.Main loop in which will read the image from canvas and do inference once.</p>
+ <textarea class="code" rows="16" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Load labels from txt file and process it into an array.</p>
+ <textarea class="code" rows="7" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+ <p>6.The post-processing, including get boxes from output and draw boxes into the image.</p>
+ <textarea class="code" rows="35" cols="100" id="codeEditor5" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [300, 300];
+mean = [127.5, 127.5, 127.5];
+std = 0.007843;
+swapRB = false;
+confThreshold = 0.5;
+nmsThreshold = 0.4;
+
+// The type of output, can be YOLO or SSD
+outType = "SSD";
+
+// url for label file, can from local or Internet
+labelsUrl = "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/object_detection_classes_pascal_voc.txt";
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+main = async function() {
+ const labels = await loadLables(labelsUrl);
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, 'canvasInput');
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const output = postProcess(result, labels);
+
+ updateResult(output, time);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+</script>
+
+<script id="codeSnippet5" type="text/code-snippet">
+postProcess = function(result, labels) {
+ let canvasOutput = document.getElementById('canvasOutput');
+ const outputWidth = canvasOutput.width;
+ const outputHeight = canvasOutput.height;
+ const resultData = result.data32F;
+
+ // Get the boxes(with class and confidence) from the output
+ let boxes = [];
+ switch(outType) {
+ case "YOLO": {
+ const vecNum = result.matSize[0];
+ const vecLength = result.matSize[1];
+ const classNum = vecLength - 5;
+
+ for (let i = 0; i < vecNum; ++i) {
+ let vector = resultData.slice(i*vecLength, (i+1)*vecLength);
+ let scores = vector.slice(5, vecLength);
+ let classId = scores.indexOf(Math.max(...scores));
+ let confidence = scores[classId];
+ if (confidence > confThreshold) {
+ let center_x = Math.round(vector[0] * outputWidth);
+ let center_y = Math.round(vector[1] * outputHeight);
+ let width = Math.round(vector[2] * outputWidth);
+ let height = Math.round(vector[3] * outputHeight);
+ let left = Math.round(center_x - width / 2);
+ let top = Math.round(center_y - height / 2);
+
+ let box = {
+ scores: scores,
+ classId: classId,
+ confidence: confidence,
+ bounding: [left, top, width, height],
+ toDraw: true
+ }
+ boxes.push(box);
+ }
+ }
+
+ // NMS(Non Maximum Suppression) algorithm
+ let boxNum = boxes.length;
+ let tmp_boxes = [];
+ let sorted_boxes = [];
+ for (let c = 0; c < classNum; ++c) {
+ for (let i = 0; i < boxes.length; ++i) {
+ tmp_boxes[i] = [boxes[i], i];
+ }
+ sorted_boxes = tmp_boxes.sort((a, b) => { return (b[0].scores[c] - a[0].scores[c]); });
+ for (let i = 0; i < boxNum; ++i) {
+ if (sorted_boxes[i][0].scores[c] === 0) continue;
+ else {
+ for (let j = i + 1; j < boxNum; ++j) {
+ if (IOU(sorted_boxes[i][0], sorted_boxes[j][0]) >= nmsThreshold) {
+ boxes[sorted_boxes[j][1]].toDraw = false;
+ }
+ }
+ }
+ }
+ }
+ } break;
+ case "SSD": {
+ const vecNum = result.matSize[2];
+ const vecLength = 7;
+
+ for (let i = 0; i < vecNum; ++i) {
+ let vector = resultData.slice(i*vecLength, (i+1)*vecLength);
+ let confidence = vector[2];
+ if (confidence > confThreshold) {
+ let left, top, right, bottom, width, height;
+ left = Math.round(vector[3]);
+ top = Math.round(vector[4]);
+ right = Math.round(vector[5]);
+ bottom = Math.round(vector[6]);
+ width = right - left + 1;
+ height = bottom - top + 1;
+ if (width <= 2 || height <= 2) {
+ left = Math.round(vector[3] * outputWidth);
+ top = Math.round(vector[4] * outputHeight);
+ right = Math.round(vector[5] * outputWidth);
+ bottom = Math.round(vector[6] * outputHeight);
+ width = right - left + 1;
+ height = bottom - top + 1;
+ }
+ let box = {
+ classId: vector[1] - 1,
+ confidence: confidence,
+ bounding: [left, top, width, height],
+ toDraw: true
+ }
+ boxes.push(box);
+ }
+ }
+ } break;
+ default:
+ console.error(`Unsupported output type ${outType}`)
+ }
+
+ // Draw the saved box into the image
+ let image = cv.imread("canvasInput");
+ let output = new cv.Mat(outputWidth, outputHeight, cv.CV_8UC3);
+ cv.cvtColor(image, output, cv.COLOR_RGBA2RGB);
+ let boxNum = boxes.length;
+ for (let i = 0; i < boxNum; ++i) {
+ if (boxes[i].toDraw) {
+ drawBox(boxes[i]);
+ }
+ }
+
+ return output;
+
+
+ // Calculate the IOU(Intersection over Union) of two boxes
+ function IOU(box1, box2) {
+ let bounding1 = box1.bounding;
+ let bounding2 = box2.bounding;
+ let s1 = bounding1[2] * bounding1[3];
+ let s2 = bounding2[2] * bounding2[3];
+
+ let left1 = bounding1[0];
+ let right1 = left1 + bounding1[2];
+ let left2 = bounding2[0];
+ let right2 = left2 + bounding2[2];
+ let overlapW = calOverlap([left1, right1], [left2, right2]);
+
+ let top1 = bounding2[1];
+ let bottom1 = top1 + bounding1[3];
+ let top2 = bounding2[1];
+ let bottom2 = top2 + bounding2[3];
+ let overlapH = calOverlap([top1, bottom1], [top2, bottom2]);
+
+ let overlapS = overlapW * overlapH;
+ return overlapS / (s1 + s2 + overlapS);
+ }
+
+ // Calculate the overlap range of two vector
+ function calOverlap(range1, range2) {
+ let min1 = range1[0];
+ let max1 = range1[1];
+ let min2 = range2[0];
+ let max2 = range2[1];
+
+ if (min2 > min1 && min2 < max1) {
+ return max1 - min2;
+ } else if (max2 > min1 && max2 < max1) {
+ return max2 - min1;
+ } else {
+ return 0;
+ }
+ }
+
+ // Draw one predict box into the origin image
+ function drawBox(box) {
+ let bounding = box.bounding;
+ let left = bounding[0];
+ let top = bounding[1];
+ let width = bounding[2];
+ let height = bounding[3];
+
+ cv.rectangle(output, new cv.Point(left, top), new cv.Point(left + width, top + height),
+ new cv.Scalar(0, 255, 0));
+ cv.rectangle(output, new cv.Point(left, top), new cv.Point(left + width, top + 15),
+ new cv.Scalar(255, 255, 255), cv.FILLED);
+ let text = `${labels[box.classId]}: ${box.confidence.toFixed(4)}`;
+ cv.putText(output, text, new cv.Point(left, top + 10), cv.FONT_HERSHEY_SIMPLEX, 0.3,
+ new cv.Scalar(0, 0, 0));
+ }
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_object_detection_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let loadLablesCode = 'loadLables = ' + loadLables.toString();
+ document.getElementById('codeEditor2').value = loadLablesCode;
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor3').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor4').value = loadModelCode;
+
+ utils.loadCode('codeSnippet5', 'codeEditor5');
+
+ let canvas = document.getElementById('canvasInput');
+ let ctx = canvas.getContext('2d');
+ let img = new Image();
+ img.crossOrigin = 'anonymous';
+ img.src = 'lena.png';
+ img.onload = function() {
+ ctx.drawImage(img, 0, 0, canvas.width, canvas.height);
+ };
+
+ let tryIt = document.getElementById('tryIt');
+ tryIt.addEventListener('click', () => {
+ initStatus();
+ document.getElementById('status').innerHTML = 'Running function main()...';
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ if (modelPath === "") {
+ document.getElementById('status').innerHTML = 'Runing failed.';
+ utils.printError('Please upload model file by clicking the button first.');
+ } else {
+ setTimeout(main, 1);
+ }
+ });
+
+ let fileInput = document.getElementById('fileInput');
+ fileInput.addEventListener('change', (e) => {
+ initStatus();
+ loadImageToCanvas(e, 'canvasInput');
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ tryIt.removeAttribute('disabled');
+ });
+
+ var main = async function() {};
+ var postProcess = function(result, labels) {};
+
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+ utils.executeCode('codeEditor5');
+
+
+ function updateResult(output, time) {
+ try{
+ let canvasOutput = document.getElementById('canvasOutput');
+ canvasOutput.style.visibility = "visible";
+ cv.imshow('canvasOutput', output);
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('canvasOutput').style.visibility = "hidden";
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+{
+ "caffe": [
+ {
+ "model": "mobilenet_SSD",
+ "inputSize": "300, 300",
+ "mean": "127.5, 127.5, 127.5",
+ "std": "0.007843",
+ "swapRB": "false",
+ "outType": "SSD",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/object_detection_classes_pascal_voc.txt",
+ "modelUrl": "https://raw.githubusercontent.com/chuanqi305/MobileNet-SSD/master/mobilenet_iter_73000.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/chuanqi305/MobileNet-SSD/master/deploy.prototxt"
+ },
+ {
+ "model": "VGG_SSD",
+ "inputSize": "300, 300",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "outType": "SSD",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/object_detection_classes_pascal_voc.txt",
+ "modelUrl": "https://drive.google.com/uc?id=0BzKzrI_SkD1_WVVTSmQxU0dVRzA&export=download",
+ "configUrl": "https://drive.google.com/uc?id=0BzKzrI_SkD1_WVVTSmQxU0dVRzA&export=download"
+ }
+ ],
+ "darknet": [
+ {
+ "model": "yolov2_tiny",
+ "inputSize": "416, 416",
+ "mean": "0, 0, 0",
+ "std": "0.00392",
+ "swapRB": "false",
+ "outType": "YOLO",
+ "labelsUrl": "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/object_detection_classes_yolov3.txt",
+ "modelUrl": "https://pjreddie.com/media/files/yolov2-tiny.weights",
+ "configUrl": "https://raw.githubusercontent.com/pjreddie/darknet/master/cfg/yolov2-tiny.cfg"
+ }
+ ]
+}
\ No newline at end of file
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Object Detection Example with Camera</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Object Detection Example with Camera </h2>
+<p>
+ This tutorial shows you how to write an object detection example with camera.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configInput</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Start/Stop</b> button to start or stop the camera capture.<br>
+</p>
+
+<div class="control"><button id="startAndStop" disabled>Start</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <video id="videoInput" width="400" height="400"></video>
+ </td>
+ <td>
+ <canvas id="canvasOutput" style="visibility: hidden;" width="400" height="400"></canvas>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ videoInput
+ </div>
+ </td>
+ <td>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile" name="file">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="15" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.The function to capture video from camera, and the main loop in which will do inference once.</p>
+ <textarea class="code" rows="34" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Load labels from txt file and process it into an array.</p>
+ <textarea class="code" rows="7" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+ <p>6.The post-processing, including get boxes from output and draw boxes into the image.</p>
+ <textarea class="code" rows="35" cols="100" id="codeEditor5" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [300, 300];
+mean = [127.5, 127.5, 127.5];
+std = 0.007843;
+swapRB = false;
+confThreshold = 0.5;
+nmsThreshold = 0.4;
+
+// the type of output, can be YOLO or SSD
+outType = "SSD";
+
+// url for label file, can from local or Internet
+labelsUrl = "https://raw.githubusercontent.com/opencv/opencv/master/samples/data/dnn/object_detection_classes_pascal_voc.txt";
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+let frame = new cv.Mat(videoInput.height, videoInput.width, cv.CV_8UC4);
+let cap = new cv.VideoCapture(videoInput);
+
+main = async function(frame) {
+ const labels = await loadLables(labelsUrl);
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, frame);
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const output = postProcess(result, labels, frame);
+
+ updateResult(output, time);
+ setTimeout(processVideo, 0);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+
+function processVideo() {
+ try {
+ if (!streaming) {
+ return;
+ }
+ cap.read(frame);
+ main(frame);
+ } catch (err) {
+ utils.printError(err);
+ }
+}
+
+setTimeout(processVideo, 0);
+</script>
+
+<script id="codeSnippet5" type="text/code-snippet">
+postProcess = function(result, labels, frame) {
+ let canvasOutput = document.getElementById('canvasOutput');
+ const outputWidth = canvasOutput.width;
+ const outputHeight = canvasOutput.height;
+ const resultData = result.data32F;
+
+ // Get the boxes(with class and confidence) from the output
+ let boxes = [];
+ switch(outType) {
+ case "YOLO": {
+ const vecNum = result.matSize[0];
+ const vecLength = result.matSize[1];
+ const classNum = vecLength - 5;
+
+ for (let i = 0; i < vecNum; ++i) {
+ let vector = resultData.slice(i*vecLength, (i+1)*vecLength);
+ let scores = vector.slice(5, vecLength);
+ let classId = scores.indexOf(Math.max(...scores));
+ let confidence = scores[classId];
+ if (confidence > confThreshold) {
+ let center_x = Math.round(vector[0] * outputWidth);
+ let center_y = Math.round(vector[1] * outputHeight);
+ let width = Math.round(vector[2] * outputWidth);
+ let height = Math.round(vector[3] * outputHeight);
+ let left = Math.round(center_x - width / 2);
+ let top = Math.round(center_y - height / 2);
+
+ let box = {
+ scores: scores,
+ classId: classId,
+ confidence: confidence,
+ bounding: [left, top, width, height],
+ toDraw: true
+ }
+ boxes.push(box);
+ }
+ }
+
+ // NMS(Non Maximum Suppression) algorithm
+ let boxNum = boxes.length;
+ let tmp_boxes = [];
+ let sorted_boxes = [];
+ for (let c = 0; c < classNum; ++c) {
+ for (let i = 0; i < boxes.length; ++i) {
+ tmp_boxes[i] = [boxes[i], i];
+ }
+ sorted_boxes = tmp_boxes.sort((a, b) => { return (b[0].scores[c] - a[0].scores[c]); });
+ for (let i = 0; i < boxNum; ++i) {
+ if (sorted_boxes[i][0].scores[c] === 0) continue;
+ else {
+ for (let j = i + 1; j < boxNum; ++j) {
+ if (IOU(sorted_boxes[i][0], sorted_boxes[j][0]) >= nmsThreshold) {
+ boxes[sorted_boxes[j][1]].toDraw = false;
+ }
+ }
+ }
+ }
+ }
+ } break;
+ case "SSD": {
+ const vecNum = result.matSize[2];
+ const vecLength = 7;
+
+ for (let i = 0; i < vecNum; ++i) {
+ let vector = resultData.slice(i*vecLength, (i+1)*vecLength);
+ let confidence = vector[2];
+ if (confidence > confThreshold) {
+ let left, top, right, bottom, width, height;
+ left = Math.round(vector[3]);
+ top = Math.round(vector[4]);
+ right = Math.round(vector[5]);
+ bottom = Math.round(vector[6]);
+ width = right - left + 1;
+ height = bottom - top + 1;
+ if (width <= 2 || height <= 2) {
+ left = Math.round(vector[3] * outputWidth);
+ top = Math.round(vector[4] * outputHeight);
+ right = Math.round(vector[5] * outputWidth);
+ bottom = Math.round(vector[6] * outputHeight);
+ width = right - left + 1;
+ height = bottom - top + 1;
+ }
+ let box = {
+ classId: vector[1] - 1,
+ confidence: confidence,
+ bounding: [left, top, width, height],
+ toDraw: true
+ }
+ boxes.push(box);
+ }
+ }
+ } break;
+ default:
+ console.error(`Unsupported output type ${outType}`)
+ }
+
+ // Draw the saved box into the image
+ let output = new cv.Mat(outputWidth, outputHeight, cv.CV_8UC3);
+ cv.cvtColor(frame, output, cv.COLOR_RGBA2RGB);
+ let boxNum = boxes.length;
+ for (let i = 0; i < boxNum; ++i) {
+ if (boxes[i].toDraw) {
+ drawBox(boxes[i]);
+ }
+ }
+
+ return output;
+
+
+ // Calculate the IOU(Intersection over Union) of two boxes
+ function IOU(box1, box2) {
+ let bounding1 = box1.bounding;
+ let bounding2 = box2.bounding;
+ let s1 = bounding1[2] * bounding1[3];
+ let s2 = bounding2[2] * bounding2[3];
+
+ let left1 = bounding1[0];
+ let right1 = left1 + bounding1[2];
+ let left2 = bounding2[0];
+ let right2 = left2 + bounding2[2];
+ let overlapW = calOverlap([left1, right1], [left2, right2]);
+
+ let top1 = bounding2[1];
+ let bottom1 = top1 + bounding1[3];
+ let top2 = bounding2[1];
+ let bottom2 = top2 + bounding2[3];
+ let overlapH = calOverlap([top1, bottom1], [top2, bottom2]);
+
+ let overlapS = overlapW * overlapH;
+ return overlapS / (s1 + s2 + overlapS);
+ }
+
+ // Calculate the overlap range of two vector
+ function calOverlap(range1, range2) {
+ let min1 = range1[0];
+ let max1 = range1[1];
+ let min2 = range2[0];
+ let max2 = range2[1];
+
+ if (min2 > min1 && min2 < max1) {
+ return max1 - min2;
+ } else if (max2 > min1 && max2 < max1) {
+ return max2 - min1;
+ } else {
+ return 0;
+ }
+ }
+
+ // Draw one predict box into the origin image
+ function drawBox(box) {
+ let bounding = box.bounding;
+ let left = bounding[0];
+ let top = bounding[1];
+ let width = bounding[2];
+ let height = bounding[3];
+
+ cv.rectangle(output, new cv.Point(left, top), new cv.Point(left + width, top + height),
+ new cv.Scalar(0, 255, 0));
+ cv.rectangle(output, new cv.Point(left, top), new cv.Point(left + width, top + 15),
+ new cv.Scalar(255, 255, 255), cv.FILLED);
+ let text = `${labels[box.classId]}: ${box.confidence.toFixed(4)}`;
+ cv.putText(output, text, new cv.Point(left, top + 10), cv.FONT_HERSHEY_SIMPLEX, 0.3,
+ new cv.Scalar(0, 0, 0));
+ }
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_object_detection_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let loadLablesCode = 'loadLables = ' + loadLables.toString();
+ document.getElementById('codeEditor2').value = loadLablesCode;
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor3').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor4').value = loadModelCode;
+
+ utils.loadCode('codeSnippet5', 'codeEditor5');
+
+ let videoInput = document.getElementById('videoInput');
+ let streaming = false;
+ let startAndStop = document.getElementById('startAndStop');
+ startAndStop.addEventListener('click', () => {
+ if (!streaming) {
+ utils.clearError();
+ utils.startCamera('qvga', onVideoStarted, 'videoInput');
+ } else {
+ utils.stopCamera();
+ onVideoStopped();
+ }
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ startAndStop.removeAttribute('disabled');
+ });
+
+ var main = async function(frame) {};
+ var postProcess = function(result, labels, frame) {};
+
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+ utils.executeCode('codeEditor5');
+
+ function onVideoStarted() {
+ streaming = true;
+ startAndStop.innerText = 'Stop';
+ videoInput.width = videoInput.videoWidth;
+ videoInput.height = videoInput.videoHeight;
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ }
+
+ function onVideoStopped() {
+ streaming = false;
+ startAndStop.innerText = 'Start';
+ initStatus();
+ }
+
+ function updateResult(output, time) {
+ try{
+ let canvasOutput = document.getElementById('canvasOutput');
+ canvasOutput.style.visibility = "visible";
+ cv.imshow('canvasOutput', output);
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('canvasOutput').style.visibility = "hidden";
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Pose Estimation Example</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Pose Estimation Example</h2>
+<p>
+ This tutorial shows you how to write an pose estimation example with OpenCV.js.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configInput</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Try it</b> button to see the result. You can choose any other images.<br>
+</p>
+
+<div class="control"><button id="tryIt" disabled>Try it</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <canvas id="canvasInput" width="400" height="250"></canvas>
+ </td>
+ <td>
+ <canvas id="canvasOutput" style="visibility: hidden;" width="400" height="250"></canvas>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ canvasInput <input type="file" id="fileInput" name="file" accept="image/*">
+ </div>
+ </td>
+ <td>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile" name="file">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="9" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.Main loop in which will read the image from canvas and do inference once.</p>
+ <textarea class="code" rows="15" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.The pairs of keypoints of different dataset.</p>
+ <textarea class="code" rows="30" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+ <p>6.The post-processing, including get the predicted points and draw lines into the image.</p>
+ <textarea class="code" rows="30" cols="100" id="codeEditor5" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [368, 368];
+mean = [0, 0, 0];
+std = 0.00392;
+swapRB = false;
+threshold = 0.1;
+
+// the pairs of keypoint, can be "COCO", "MPI" and "BODY_25"
+dataset = "COCO";
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+main = async function() {
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, 'canvasInput');
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const output = postProcess(result);
+
+ updateResult(output, time);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+</script>
+
+<script id="codeSnippet4" type="text/code-snippet">
+BODY_PARTS = {};
+POSE_PAIRS = [];
+
+if (dataset === 'COCO') {
+ BODY_PARTS = { "Nose": 0, "Neck": 1, "RShoulder": 2, "RElbow": 3, "RWrist": 4,
+ "LShoulder": 5, "LElbow": 6, "LWrist": 7, "RHip": 8, "RKnee": 9,
+ "RAnkle": 10, "LHip": 11, "LKnee": 12, "LAnkle": 13, "REye": 14,
+ "LEye": 15, "REar": 16, "LEar": 17, "Background": 18 };
+
+ POSE_PAIRS = [ ["Neck", "RShoulder"], ["Neck", "LShoulder"], ["RShoulder", "RElbow"],
+ ["RElbow", "RWrist"], ["LShoulder", "LElbow"], ["LElbow", "LWrist"],
+ ["Neck", "RHip"], ["RHip", "RKnee"], ["RKnee", "RAnkle"], ["Neck", "LHip"],
+ ["LHip", "LKnee"], ["LKnee", "LAnkle"], ["Neck", "Nose"], ["Nose", "REye"],
+ ["REye", "REar"], ["Nose", "LEye"], ["LEye", "LEar"] ]
+} else if (dataset === 'MPI') {
+ BODY_PARTS = { "Head": 0, "Neck": 1, "RShoulder": 2, "RElbow": 3, "RWrist": 4,
+ "LShoulder": 5, "LElbow": 6, "LWrist": 7, "RHip": 8, "RKnee": 9,
+ "RAnkle": 10, "LHip": 11, "LKnee": 12, "LAnkle": 13, "Chest": 14,
+ "Background": 15 }
+
+ POSE_PAIRS = [ ["Head", "Neck"], ["Neck", "RShoulder"], ["RShoulder", "RElbow"],
+ ["RElbow", "RWrist"], ["Neck", "LShoulder"], ["LShoulder", "LElbow"],
+ ["LElbow", "LWrist"], ["Neck", "Chest"], ["Chest", "RHip"], ["RHip", "RKnee"],
+ ["RKnee", "RAnkle"], ["Chest", "LHip"], ["LHip", "LKnee"], ["LKnee", "LAnkle"] ]
+} else if (dataset === 'BODY_25') {
+ BODY_PARTS = { "Nose": 0, "Neck": 1, "RShoulder": 2, "RElbow": 3, "RWrist": 4,
+ "LShoulder": 5, "LElbow": 6, "LWrist": 7, "MidHip": 8, "RHip": 9,
+ "RKnee": 10, "RAnkle": 11, "LHip": 12, "LKnee": 13, "LAnkle": 14,
+ "REye": 15, "LEye": 16, "REar": 17, "LEar": 18, "LBigToe": 19,
+ "LSmallToe": 20, "LHeel": 21, "RBigToe": 22, "RSmallToe": 23,
+ "RHeel": 24, "Background": 25 }
+
+ POSE_PAIRS = [ ["Neck", "Nose"], ["Neck", "RShoulder"],
+ ["Neck", "LShoulder"], ["RShoulder", "RElbow"],
+ ["RElbow", "RWrist"], ["LShoulder", "LElbow"],
+ ["LElbow", "LWrist"], ["Nose", "REye"],
+ ["REye", "REar"], ["Neck", "LEye"],
+ ["LEye", "LEar"], ["Neck", "MidHip"],
+ ["MidHip", "RHip"], ["RHip", "RKnee"],
+ ["RKnee", "RAnkle"], ["RAnkle", "RBigToe"],
+ ["RBigToe", "RSmallToe"], ["RAnkle", "RHeel"],
+ ["MidHip", "LHip"], ["LHip", "LKnee"],
+ ["LKnee", "LAnkle"], ["LAnkle", "LBigToe"],
+ ["LBigToe", "LSmallToe"], ["LAnkle", "LHeel"] ]
+}
+</script>
+
+<script id="codeSnippet5" type="text/code-snippet">
+postProcess = function(result) {
+ const resultData = result.data32F;
+ const matSize = result.matSize;
+ const size1 = matSize[1];
+ const size2 = matSize[2];
+ const size3 = matSize[3];
+ const mapSize = size2 * size3;
+
+ let canvasOutput = document.getElementById('canvasOutput');
+ const outputWidth = canvasOutput.width;
+ const outputHeight = canvasOutput.height;
+
+ let image = cv.imread("canvasInput");
+ let output = new cv.Mat(outputWidth, outputHeight, cv.CV_8UC3);
+ cv.cvtColor(image, output, cv.COLOR_RGBA2RGB);
+
+ // get position of keypoints from output
+ let points = [];
+ for (let i = 0; i < Object.keys(BODY_PARTS).length; ++i) {
+ heatMap = resultData.slice(i*mapSize, (i+1)*mapSize);
+
+ let maxIndex = 0;
+ let maxConf = heatMap[0];
+ for (index in heatMap) {
+ if (heatMap[index] > heatMap[maxIndex]) {
+ maxIndex = index;
+ maxConf = heatMap[index];
+ }
+ }
+
+ if (maxConf > threshold) {
+ indexX = maxIndex % size3;
+ indexY = maxIndex / size3;
+
+ x = outputWidth * indexX / size3;
+ y = outputHeight * indexY / size2;
+
+ points[i] = [Math.round(x), Math.round(y)];
+ }
+ }
+
+ // draw the points and lines into the image
+ for (pair of POSE_PAIRS) {
+ partFrom = pair[0];
+ partTo = pair[1];
+ idFrom = BODY_PARTS[partFrom];
+ idTo = BODY_PARTS[partTo];
+ pointFrom = points[idFrom];
+ pointTo = points[idTo];
+
+ if (points[idFrom] && points[idTo]) {
+ cv.line(output, new cv.Point(pointFrom[0], pointFrom[1]),
+ new cv.Point(pointTo[0], pointTo[1]), new cv.Scalar(0, 255, 0), 3);
+ cv.ellipse(output, new cv.Point(pointFrom[0], pointFrom[1]), new cv.Size(3, 3), 0, 0, 360,
+ new cv.Scalar(0, 0, 255), cv.FILLED);
+ cv.ellipse(output, new cv.Point(pointTo[0], pointTo[1]), new cv.Size(3, 3), 0, 0, 360,
+ new cv.Scalar(0, 0, 255), cv.FILLED);
+ }
+ }
+
+ return output;
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_pose_estimation_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor2').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor3').value = loadModelCode;
+
+ utils.loadCode('codeSnippet4', 'codeEditor4');
+ utils.loadCode('codeSnippet5', 'codeEditor5');
+
+ let canvas = document.getElementById('canvasInput');
+ let ctx = canvas.getContext('2d');
+ let img = new Image();
+ img.crossOrigin = 'anonymous';
+ img.src = 'roi.jpg';
+ img.onload = function() {
+ ctx.drawImage(img, 0, 0, canvas.width, canvas.height);
+ };
+
+ let tryIt = document.getElementById('tryIt');
+ tryIt.addEventListener('click', () => {
+ initStatus();
+ document.getElementById('status').innerHTML = 'Running function main()...';
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ if (modelPath === "") {
+ document.getElementById('status').innerHTML = 'Runing failed.';
+ utils.printError('Please upload model file by clicking the button first.');
+ } else {
+ setTimeout(main, 1);
+ }
+ });
+
+ let fileInput = document.getElementById('fileInput');
+ fileInput.addEventListener('change', (e) => {
+ initStatus();
+ loadImageToCanvas(e, 'canvasInput');
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ tryIt.removeAttribute('disabled');
+ });
+
+ var main = async function() {};
+ var postProcess = function(result) {};
+
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+ utils.executeCode('codeEditor5');
+
+ function updateResult(output, time) {
+ try{
+ let canvasOutput = document.getElementById('canvasOutput');
+ canvasOutput.style.visibility = "visible";
+ let resized = new cv.Mat(canvasOutput.width, canvasOutput.height, cv.CV_8UC4);
+ cv.resize(output, resized, new cv.Size(canvasOutput.width, canvasOutput.height));
+ cv.imshow('canvasOutput', resized);
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('canvasOutput').style.visibility = "hidden";
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+{
+ "caffe": [
+ {
+ "model": "body_25",
+ "inputSize": "368, 368",
+ "mean": "0, 0, 0",
+ "std": "0.00392",
+ "swapRB": "false",
+ "dataset": "BODY_25",
+ "modelUrl": "http://posefs1.perception.cs.cmu.edu/OpenPose/models/pose/body_25/pose_iter_584000.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/CMU-Perceptual-Computing-Lab/openpose/master/models/pose/body_25/pose_deploy.prototxt"
+ },
+ {
+ "model": "coco",
+ "inputSize": "368, 368",
+ "mean": "0, 0, 0",
+ "std": "0.00392",
+ "swapRB": "false",
+ "dataset": "COCO",
+ "modelUrl": "http://posefs1.perception.cs.cmu.edu/OpenPose/models/pose/coco/pose_iter_440000.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/CMU-Perceptual-Computing-Lab/openpose/master/models/pose/coco/pose_deploy_linevec.prototxt"
+ },
+ {
+ "model": "mpi",
+ "inputSize": "368, 368",
+ "mean": "0, 0, 0",
+ "std": "0.00392",
+ "swapRB": "false",
+ "dataset": "MPI",
+ "modelUrl": "http://posefs1.perception.cs.cmu.edu/OpenPose/models/pose/mpi/pose_iter_160000.caffemodel",
+ "configUrl": "https://raw.githubusercontent.com/CMU-Perceptual-Computing-Lab/openpose/master/models/pose/mpi/pose_deploy_linevec.prototxt"
+ }
+ ]
+}
\ No newline at end of file
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Semantic Segmentation Example</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Semantic Segmentation Example</h2>
+<p>
+ This tutorial shows you how to write an semantic segmentation example with OpenCV.js.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configInput</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Try it</b> button to see the result. You can choose any other images.<br>
+</p>
+
+<div class="control"><button id="tryIt" disabled>Try it</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <canvas id="canvasInput" width="400" height="400"></canvas>
+ </td>
+ <td>
+ <canvas id="canvasOutput" style="visibility: hidden;" width="400" height="400"></canvas>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ canvasInput <input type="file" id="fileInput" name="file" accept="image/*">
+ </div>
+ </td>
+ <td>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile" name="file">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="5" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.Main loop in which will read the image from canvas and do inference once.</p>
+ <textarea class="code" rows="16" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.The post-processing, including gengerate colors for different classes and argmax to get the classes for each pixel.</p>
+ <textarea class="code" rows="34" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [513, 513];
+mean = [127.5, 127.5, 127.5];
+std = 0.007843;
+swapRB = false;
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+main = async function() {
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, 'canvasInput');
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const colors = generateColors(result);
+ const output = argmax(result, colors);
+
+ updateResult(output, time);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+</script>
+
+<script id="codeSnippet4" type="text/code-snippet">
+generateColors = function(result) {
+ const numClasses = result.matSize[1];
+ let colors = [0,0,0];
+ while(colors.length < numClasses*3){
+ colors.push(Math.round((Math.random()*255 + colors[colors.length-3]) / 2));
+ }
+ return colors;
+}
+
+argmax = function(result, colors) {
+ const C = result.matSize[1];
+ const H = result.matSize[2];
+ const W = result.matSize[3];
+ const resultData = result.data32F;
+ const imgSize = H*W;
+
+ let classId = [];
+ for (i = 0; i<imgSize; ++i) {
+ let id = 0;
+ for (j = 0; j < C; ++j) {
+ if (resultData[j*imgSize+i] > resultData[id*imgSize+i]) {
+ id = j;
+ }
+ }
+ classId.push(colors[id*3]);
+ classId.push(colors[id*3+1]);
+ classId.push(colors[id*3+2]);
+ classId.push(255);
+ }
+
+ output = cv.matFromArray(H,W,cv.CV_8UC4,classId);
+ return output;
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_semantic_segmentation_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor2').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor3').value = loadModelCode;
+
+ utils.loadCode('codeSnippet4', 'codeEditor4');
+
+ let canvas = document.getElementById('canvasInput');
+ let ctx = canvas.getContext('2d');
+ let img = new Image();
+ img.crossOrigin = 'anonymous';
+ img.src = 'roi.jpg';
+ img.onload = function() {
+ ctx.drawImage(img, 0, 0, canvas.width, canvas.height);
+ };
+
+ let tryIt = document.getElementById('tryIt');
+ tryIt.addEventListener('click', () => {
+ initStatus();
+ document.getElementById('status').innerHTML = 'Running function main()...';
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ if (modelPath === "") {
+ document.getElementById('status').innerHTML = 'Runing failed.';
+ utils.printError('Please upload model file by clicking the button first.');
+ } else {
+ setTimeout(main, 1);
+ }
+ });
+
+ let fileInput = document.getElementById('fileInput');
+ fileInput.addEventListener('change', (e) => {
+ initStatus();
+ loadImageToCanvas(e, 'canvasInput');
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ tryIt.removeAttribute('disabled');
+ });
+
+ var main = async function() {};
+ var generateColors = function(result) {};
+ var argmax = function(result, colors) {};
+
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+
+ function updateResult(output, time) {
+ try{
+ let canvasOutput = document.getElementById('canvasOutput');
+ canvasOutput.style.visibility = "visible";
+ let resized = new cv.Mat(canvasOutput.width, canvasOutput.height, cv.CV_8UC4);
+ cv.resize(output, resized, new cv.Size(canvasOutput.width, canvasOutput.height));
+ cv.imshow('canvasOutput', resized);
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('canvasOutput').style.visibility = "hidden";
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+{
+ "tensorflow": [
+ {
+ "model": "deeplabv3",
+ "inputSize": "513, 513",
+ "mean": "127.5, 127.5, 127.5",
+ "std": "0.007843",
+ "swapRB": "false",
+ "modelUrl": "https://drive.google.com/uc?id=1v-hfGenaE9tiGOzo5qdgMNG_gqQ5-Xn4&export=download"
+ }
+ ]
+}
\ No newline at end of file
--- /dev/null
+<!DOCTYPE html>
+<html>
+
+<head>
+ <meta charset="utf-8">
+ <title>Style Transfer Example</title>
+ <link href="js_example_style.css" rel="stylesheet" type="text/css" />
+</head>
+
+<body>
+<h2>Style Transfer Example</h2>
+<p>
+ This tutorial shows you how to write an style transfer example with OpenCV.js.<br>
+ To try the example you should click the <b>modelFile</b> button(and <b>configFile</b> button if needed) to upload inference model.
+ You can find the model URLs and parameters in the <a href="#appendix">model info</a> section.
+ Then You should change the parameters in the first code snippet according to the uploaded model.
+ Finally click <b>Try it</b> button to see the result. You can choose any other images.<br>
+</p>
+
+<div class="control"><button id="tryIt" disabled>Try it</button></div>
+<div>
+ <table cellpadding="0" cellspacing="0" width="0" border="0">
+ <tr>
+ <td>
+ <canvas id="canvasInput" width="400" height="400"></canvas>
+ </td>
+ <td>
+ <canvas id="canvasOutput" style="visibility: hidden;" width="400" height="400"></canvas>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ canvasInput <input type="file" id="fileInput" name="file" accept="image/*">
+ </div>
+ </td>
+ <td>
+ <p id='status' align="left"></p>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ modelFile <input type="file" id="modelFile" name="file">
+ </div>
+ </td>
+ </tr>
+ <tr>
+ <td>
+ <div class="caption">
+ configFile <input type="file" id="configFile">
+ </div>
+ </td>
+ </tr>
+ </table>
+</div>
+
+<div>
+ <p class="err" id="errorMessage"></p>
+</div>
+
+<div>
+ <h3>Help function</h3>
+ <p>1.The parameters for model inference which you can modify to investigate more models.</p>
+ <textarea class="code" rows="5" cols="100" id="codeEditor" spellcheck="false"></textarea>
+ <p>2.Main loop in which will read the image from canvas and do inference once.</p>
+ <textarea class="code" rows="15" cols="100" id="codeEditor1" spellcheck="false"></textarea>
+ <p>3.Get blob from image as input for net, and standardize it with <b>mean</b> and <b>std</b>.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor2" spellcheck="false"></textarea>
+ <p>4.Fetch model file and save to emscripten file system once click the input button.</p>
+ <textarea class="code" rows="17" cols="100" id="codeEditor3" spellcheck="false"></textarea>
+ <p>5.The post-processing, including scaling and reordering.</p>
+ <textarea class="code" rows="21" cols="100" id="codeEditor4" spellcheck="false"></textarea>
+</div>
+
+<div id="appendix">
+ <h2>Model Info:</h2>
+</div>
+
+<script src="utils.js" type="text/javascript"></script>
+<script src="js_dnn_example_helper.js" type="text/javascript"></script>
+
+<script id="codeSnippet" type="text/code-snippet">
+inputSize = [224, 224];
+mean = [104, 117, 123];
+std = 1;
+swapRB = false;
+</script>
+
+<script id="codeSnippet1" type="text/code-snippet">
+main = async function() {
+ const input = getBlobFromImage(inputSize, mean, std, swapRB, 'canvasInput');
+ let net = cv.readNet(configPath, modelPath);
+ net.setInput(input);
+ const start = performance.now();
+ const result = net.forward();
+ const time = performance.now()-start;
+ const output = postProcess(result);
+
+ updateResult(output, time);
+ input.delete();
+ net.delete();
+ result.delete();
+}
+</script>
+
+<script id="codeSnippet4" type="text/code-snippet">
+postProcess = function(result) {
+ const resultData = result.data32F;
+ const C = result.matSize[1];
+ const H = result.matSize[2];
+ const W = result.matSize[3];
+ const mean = [104, 117, 123];
+
+ let normData = [];
+ for (let h = 0; h < H; ++h) {
+ for (let w = 0; w < W; ++w) {
+ for (let c = 0; c < C; ++c) {
+ normData.push(resultData[c*H*W + h*W + w] + mean[c]);
+ }
+ normData.push(255);
+ }
+ }
+
+ let output = new cv.matFromArray(H, W, cv.CV_8UC4, normData);
+ return output;
+}
+</script>
+
+<script type="text/javascript">
+ let jsonUrl = "js_style_transfer_model_info.json";
+ drawInfoTable(jsonUrl, 'appendix');
+
+ let utils = new Utils('errorMessage');
+ utils.loadCode('codeSnippet', 'codeEditor');
+ utils.loadCode('codeSnippet1', 'codeEditor1');
+
+ let getBlobFromImageCode = 'getBlobFromImage = ' + getBlobFromImage.toString();
+ document.getElementById('codeEditor2').value = getBlobFromImageCode;
+ let loadModelCode = 'loadModel = ' + loadModel.toString();
+ document.getElementById('codeEditor3').value = loadModelCode;
+
+ utils.loadCode('codeSnippet4', 'codeEditor4');
+
+ let canvas = document.getElementById('canvasInput');
+ let ctx = canvas.getContext('2d');
+ let img = new Image();
+ img.crossOrigin = 'anonymous';
+ img.src = 'lena.png';
+ img.onload = function() {
+ ctx.drawImage(img, 0, 0, canvas.width, canvas.height);
+ };
+
+ let tryIt = document.getElementById('tryIt');
+ tryIt.addEventListener('click', () => {
+ initStatus();
+ document.getElementById('status').innerHTML = 'Running function main()...';
+ utils.executeCode('codeEditor');
+ utils.executeCode('codeEditor1');
+ if (modelPath === "") {
+ document.getElementById('status').innerHTML = 'Runing failed.';
+ utils.printError('Please upload model file by clicking the button first.');
+ } else {
+ setTimeout(main, 1);
+ }
+ });
+
+ let fileInput = document.getElementById('fileInput');
+ fileInput.addEventListener('change', (e) => {
+ initStatus();
+ loadImageToCanvas(e, 'canvasInput');
+ });
+
+ let configPath = "";
+ let configFile = document.getElementById('configFile');
+ configFile.addEventListener('change', async (e) => {
+ initStatus();
+ configPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The config file '${configPath}' is created successfully.`;
+ });
+
+ let modelPath = "";
+ let modelFile = document.getElementById('modelFile');
+ modelFile.addEventListener('change', async (e) => {
+ initStatus();
+ modelPath = await loadModel(e);
+ document.getElementById('status').innerHTML = `The model file '${modelPath}' is created successfully.`;
+ configPath = "";
+ configFile.value = "";
+ });
+
+ utils.loadOpenCv(() => {
+ tryIt.removeAttribute('disabled');
+ });
+
+ var main = async function() {};
+ var postProcess = function(result) {};
+
+ utils.executeCode('codeEditor1');
+ utils.executeCode('codeEditor2');
+ utils.executeCode('codeEditor3');
+ utils.executeCode('codeEditor4');
+
+ function updateResult(output, time) {
+ try{
+ let canvasOutput = document.getElementById('canvasOutput');
+ canvasOutput.style.visibility = "visible";
+ let resized = new cv.Mat(canvasOutput.width, canvasOutput.height, cv.CV_8UC4);
+ cv.resize(output, resized, new cv.Size(canvasOutput.width, canvasOutput.height));
+ cv.imshow('canvasOutput', resized);
+ document.getElementById('status').innerHTML = `<b>Model:</b> ${modelPath}<br>
+ <b>Inference time:</b> ${time.toFixed(2)} ms`;
+ } catch(e) {
+ console.log(e);
+ }
+ }
+
+ function initStatus() {
+ document.getElementById('status').innerHTML = '';
+ document.getElementById('canvasOutput').style.visibility = "hidden";
+ utils.clearError();
+ }
+
+</script>
+
+</body>
+
+</html>
\ No newline at end of file
--- /dev/null
+{
+ "torch": [
+ {
+ "model": "candy.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//instance_norm/candy.t7"
+ },
+ {
+ "model": "composition_vii.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//eccv16/composition_vii.t7"
+ },
+ {
+ "model": "feathers.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//instance_norm/feathers.t7"
+ },
+ {
+ "model": "la_muse.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//instance_norm/la_muse.t7"
+ },
+ {
+ "model": "mosaic.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//instance_norm/mosaic.t7"
+ },
+ {
+ "model": "starry_night.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//eccv16/starry_night.t7"
+ },
+ {
+ "model": "the_scream.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//instance_norm/the_scream.t7"
+ },
+ {
+ "model": "the_wave.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//eccv16/the_wave.t7"
+ },
+ {
+ "model": "udnie.t7",
+ "inputSize": "224, 224",
+ "mean": "104, 117, 123",
+ "std": "1",
+ "swapRB": "false",
+ "modelUrl": "https://cs.stanford.edu/people/jcjohns/fast-neural-style/models//instance_norm/udnie.t7"
+ }
+ ]
+}
\ No newline at end of file
let script = document.createElement('script');
script.setAttribute('async', '');
script.setAttribute('type', 'text/javascript');
- script.addEventListener('load', () => {
+ script.addEventListener('load', async () => {
if (cv.getBuildInformation)
{
console.log(cv.getBuildInformation());
else
{
// WASM
- cv['onRuntimeInitialized']=()=>{
+ if (cv instanceof Promise) {
+ cv = await cv;
console.log(cv.getBuildInformation());
onloadCallback();
+ } else {
+ cv['onRuntimeInitialized']=()=>{
+ console.log(cv.getBuildInformation());
+ onloadCallback();
+ }
}
}
});
--- /dev/null
+Image Classification Example {#tutorial_js_image_classification}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for image classification.
+
+\htmlonly
+<iframe src="../../js_image_classification.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Image Classification Example with Camera {#tutorial_js_image_classification_with_camera}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for image classification example with camera.
+
+@note If you don't know how to capture video from camera, please review @ref tutorial_js_video_display.
+
+\htmlonly
+<iframe src="../../js_image_classification_with_camera.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Object Detection Example {#tutorial_js_object_detection}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for object detection.
+
+\htmlonly
+<iframe src="../../js_object_detection.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Object Detection Example with Camera{#tutorial_js_object_detection_with_camera}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for object detection with camera.
+
+\htmlonly
+<iframe src="../../js_object_detection_with_camera.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Pose Estimation Example {#tutorial_js_pose_estimation}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for pose estimation.
+
+\htmlonly
+<iframe src="../../js_pose_estimation.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Semantic Segmentation Example {#tutorial_js_semantic_segmentation}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for semantic segmentation.
+
+\htmlonly
+<iframe src="../../js_semantic_segmentation.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Style Transfer Example {#tutorial_js_style_transfer}
+=======================================
+
+Goal
+----
+
+- In this tutorial you will learn how to use OpenCV.js dnn module for style transfer.
+
+\htmlonly
+<iframe src="../../js_style_transfer.html" width="100%"
+ onload="this.style.height=this.contentDocument.body.scrollHeight +'px';">
+</iframe>
+\endhtmlonly
\ No newline at end of file
--- /dev/null
+Deep Neural Networks (dnn module) {#tutorial_js_table_of_contents_dnn}
+============
+
+- @subpage tutorial_js_image_classification
+
+ Image classification example
+
+- @subpage tutorial_js_image_classification_with_camera
+
+ Image classification example with camera
+
+- @subpage tutorial_js_object_detection
+
+ Object detection example
+
+- @subpage tutorial_js_object_detection_with_camera
+
+ Object detection example with camera
+
+- @subpage tutorial_js_semantic_segmentation
+
+ Semantic segmentation example
+
+- @subpage tutorial_js_style_transfer
+
+ Style transfer example
+
+- @subpage tutorial_js_pose_estimation
+
+ Pose estimation example
In this section you
will object detection techniques like face detection etc.
+
+- @subpage tutorial_js_table_of_contents_dnn
+
+ These tutorials show how to use dnn module in JavaScript
@code{.cpp}
CV_FOURCC('P','I','M,'1') // this is an MPEG1 codec from the characters to integer
@endcode
- If you pass for this argument minus one than a window will pop up at runtime that contains all
+ If you pass for this argument minus one then a window will pop up at runtime that contains all
the codec installed on your system and ask you to select the one to use:
The joint rotation-translation matrix \f$[R|t]\f$ is the matrix product of a projective
transformation and a homogeneous transformation. The 3-by-4 projective transformation maps 3D points
-represented in camera coordinates to 2D poins in the image plane and represented in normalized
+represented in camera coordinates to 2D points in the image plane and represented in normalized
camera coordinates \f$x' = X_c / Z_c\f$ and \f$y' = Y_c / Z_c\f$:
\f[Z_c \begin{bmatrix}
vector<Point2f> test_corners;
bool result = findChessboardCorners(chessboard_image, pattern_size, test_corners, 15);
- if(!result)
+ if (!result && cvtest::debugLevel > 0)
{
-#if 0
ts->printf(cvtest::TS::LOG, "Warning: chessboard was not detected! Writing image to test.png\n");
ts->printf(cvtest::TS::LOG, "Size = %d, %d\n", pattern_size.width, pattern_size.height);
ts->printf(cvtest::TS::LOG, "Intrinsic params: fx = %f, fy = %f, cx = %f, cy = %f\n",
distortion_coeffs_.at<double>(0, 4));
imwrite("test.png", chessboard_image);
-#endif
+ }
+ if (!result)
+ {
continue;
}
<< "Q =" << std::endl << Q << std::endl;
}
-#if 1 // Debug code
+ if (cvtest::debugLevel == 0)
+ return;
+ // DEBUG code is below
+
cv::Mat lmapx, lmapy, rmapx, rmapy;
//rewrite for fisheye
cv::fisheye::initUndistortRectifyMap(K1, D1, R1, P1, requested_size, CV_32F, lmapx, lmapy);
cv::imwrite(cv::format("fisheye_rectification_AB_%03d.png", i), rectification);
}
-#endif
}
TEST_F(fisheyeTest, stereoCalibrate)
UMat(const UMat& m, const Rect& roi);
UMat(const UMat& m, const Range* ranges);
UMat(const UMat& m, const std::vector<Range>& ranges);
+
+ // FIXIT copyData=false is not implemented, drop this in favor of cv::Mat (OpenCV 5.0)
//! builds matrix from std::vector with or without copying the data
template<typename _Tp> explicit UMat(const std::vector<_Tp>& vec, bool copyData=false);
- //! builds matrix from cv::Vec; the data is copied by default
- template<typename _Tp, int n> explicit UMat(const Vec<_Tp, n>& vec, bool copyData=true);
- //! builds matrix from cv::Matx; the data is copied by default
- template<typename _Tp, int m, int n> explicit UMat(const Matx<_Tp, m, n>& mtx, bool copyData=true);
- //! builds matrix from a 2D point
- template<typename _Tp> explicit UMat(const Point_<_Tp>& pt, bool copyData=true);
- //! builds matrix from a 3D point
- template<typename _Tp> explicit UMat(const Point3_<_Tp>& pt, bool copyData=true);
- //! builds matrix from comma initializer
- template<typename _Tp> explicit UMat(const MatCommaInitializer_<_Tp>& commaInitializer);
-
//! destructor - calls release()
~UMat();
//! assignment operators
{
CV_INSTRUMENT_REGION_IPP();
-#if defined __APPLE__ || (defined _MSC_VER && defined _M_IX86)
// see https://github.com/opencv/opencv/issues/17453
- if (src.dims <= 2 && src.step > 520000)
+ if (src.dims <= 2 && src.step > 520000 && cv::ipp::getIppTopFeatures() == ippCPUID_SSE42)
return false;
-#endif
#if IPP_VERSION_X100 < 201801
// Poor performance of SSE42
{
v_float32 t0 = vx_load(a + j) - vx_load(b + j);
v_float32 t1 = vx_load(a + j + v_float32::nlanes) - vx_load(b + j + v_float32::nlanes);
- v_float32 t2 = vx_load(a + j + 2 * v_float32::nlanes) - vx_load(b + j + 2 * v_float32::nlanes);
- v_float32 t3 = vx_load(a + j + 3 * v_float32::nlanes) - vx_load(b + j + 3 * v_float32::nlanes);
v_d0 = v_muladd(t0, t0, v_d0);
+ v_float32 t2 = vx_load(a + j + 2 * v_float32::nlanes) - vx_load(b + j + 2 * v_float32::nlanes);
v_d1 = v_muladd(t1, t1, v_d1);
+ v_float32 t3 = vx_load(a + j + 3 * v_float32::nlanes) - vx_load(b + j + 3 * v_float32::nlanes);
v_d2 = v_muladd(t2, t2, v_d2);
v_d3 = v_muladd(t3, t3, v_d3);
}
ippTopFeatures = ippCPUID_SSE42;
pIppLibInfo = ippiGetLibVersion();
+
+ // workaround: https://github.com/opencv/opencv/issues/12959
+ std::string ippName(pIppLibInfo->Name ? pIppLibInfo->Name : "");
+ if (ippName.find("SSE4.2") != std::string::npos)
+ {
+ ippTopFeatures = ippCPUID_SSE42;
+ }
}
public:
#endif
}
-unsigned long long getIppTopFeatures();
-
+#ifdef HAVE_IPP
unsigned long long getIppTopFeatures()
{
-#ifdef HAVE_IPP
return getIPPSingleton().ippTopFeatures;
-#else
- return 0;
-#endif
}
+#endif
void setIppStatus(int status, const char * const _funcname, const char * const _filename, int _line)
{
img->copyTo(sub);
shift += img->size().height + 1;
}
- //imwrite("/tmp/all_fonts.png", result);
+ if (cvtest::debugLevel > 0)
+ imwrite("all_fonts.png", result);
}
};
# ----------------------------------------------------------------------------
# CMake file for js support
# ----------------------------------------------------------------------------
-set(the_description "The js bindings")
+if(OPENCV_INITIAL_PASS)
+ # generator for Objective-C source code and documentation signatures
+ add_subdirectory(generator)
+endif()
if(NOT BUILD_opencv_js) # should be enabled explicitly (by build_js.py script)
- ocv_module_disable(js)
+ return()
endif()
+set(the_description "The JavaScript(JS) bindings")
+
set(OPENCV_JS "opencv.js")
+set(JS_HELPER "${CMAKE_CURRENT_SOURCE_DIR}/src/helpers.js")
find_path(EMSCRIPTEN_INCLUDE_DIR
emscripten/bind.h
ocv_module_disable(js)
endif()
-ocv_add_module(js BINDINGS)
+ocv_add_module(js BINDINGS PRIVATE_REQUIRED opencv_js_bindings_generator)
ocv_module_include_directories(${EMSCRIPTEN_INCLUDE_DIR})
-# get list of modules to wrap
-# message(STATUS "Wrapped in js:")
-set(OPENCV_JS_MODULES)
-foreach(m ${OPENCV_MODULES_BUILD})
- if(";${OPENCV_MODULE_${m}_WRAPPERS};" MATCHES ";js;" AND HAVE_${m})
- list(APPEND OPENCV_JS_MODULES ${m})
- # message(STATUS "\t${m}")
- endif()
-endforeach()
-
-set(opencv_hdrs "")
-foreach(m ${OPENCV_JS_MODULES})
- list(APPEND opencv_hdrs ${OPENCV_MODULE_${m}_HEADERS})
-endforeach(m)
-
-# header blacklist
-ocv_list_filterout(opencv_hdrs "modules/.*.h$")
-ocv_list_filterout(opencv_hdrs "modules/core/.*/cuda")
-ocv_list_filterout(opencv_hdrs "modules/core/.*/opencl")
-ocv_list_filterout(opencv_hdrs "modules/core/include/opencv2/core/opengl.hpp")
-ocv_list_filterout(opencv_hdrs "modules/core/include/opencv2/core/ocl.hpp")
-ocv_list_filterout(opencv_hdrs "modules/cuda.*")
-ocv_list_filterout(opencv_hdrs "modules/cudev")
-ocv_list_filterout(opencv_hdrs "modules/core/.*/hal/")
-ocv_list_filterout(opencv_hdrs "modules/.*/detection_based_tracker.hpp") # Conditional compilation
-ocv_list_filterout(opencv_hdrs "modules/core/include/opencv2/core/utils/.*")
-
-file(WRITE "${CMAKE_CURRENT_BINARY_DIR}/headers.txt" "${opencv_hdrs}")
-
-set(bindings_cpp "${CMAKE_CURRENT_BINARY_DIR}/bindings.cpp")
-
-set(scripts_hdr_parser "${CMAKE_CURRENT_SOURCE_DIR}/../python/src2/hdr_parser.py")
-
-set(JS_HELPER "${CMAKE_CURRENT_SOURCE_DIR}/src/helpers.js")
-
-add_custom_command(
- OUTPUT ${bindings_cpp}
- COMMAND ${PYTHON_DEFAULT_EXECUTABLE} "${CMAKE_CURRENT_SOURCE_DIR}/src/embindgen.py" ${scripts_hdr_parser} ${bindings_cpp} "${CMAKE_CURRENT_BINARY_DIR}/headers.txt" "${CMAKE_CURRENT_SOURCE_DIR}/src/core_bindings.cpp"
- DEPENDS ${CMAKE_CURRENT_SOURCE_DIR}/src/core_bindings.cpp
- DEPENDS ${CMAKE_CURRENT_SOURCE_DIR}/src/embindgen.py
- DEPENDS ${CMAKE_CURRENT_SOURCE_DIR}/src/templates.py
- DEPENDS ${scripts_hdr_parser}
- #(not needed - generated by CMake) DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/headers.txt
- DEPENDS ${opencv_hdrs}
- DEPENDS ${JS_HELPER})
-
add_definitions("-std=c++11")
-link_libraries(${OPENCV_MODULE_${the_module}_DEPS})
+set(deps ${OPENCV_MODULE_${the_module}_DEPS})
+list(REMOVE_ITEM deps opencv_js_bindings_generator) # don't add dummy module
+link_libraries(${deps})
+
+set(bindings_cpp "${OPENCV_JS_BINDINGS_DIR}/gen/bindings.cpp")
+set_source_files_properties(${bindings_cpp} PROPERTIES GENERATED TRUE)
OCV_OPTION(BUILD_WASM_INTRIN_TESTS "Build WASM intrin tests" OFF )
if(BUILD_WASM_INTRIN_TESTS)
ocv_add_executable(${the_module} ${bindings_cpp})
endif()
+add_dependencies(${the_module} gen_opencv_js_source)
+
set(COMPILE_FLAGS "")
if(NOT CMAKE_CXX_COMPILER_ID MATCHES "MSVC")
set(COMPILE_FLAGS "${COMPILE_FLAGS} -Wno-missing-prototypes")
if(COMPILE_FLAGS)
set_target_properties(${the_module} PROPERTIES COMPILE_FLAGS ${COMPILE_FLAGS})
endif()
+
set(EMSCRIPTEN_LINK_FLAGS "${EMSCRIPTEN_LINK_FLAGS} --memory-init-file 0 -s TOTAL_MEMORY=128MB -s WASM_MEM_MAX=1GB -s ALLOW_MEMORY_GROWTH=1")
set(EMSCRIPTEN_LINK_FLAGS "${EMSCRIPTEN_LINK_FLAGS} -s MODULARIZE=1 -s SINGLE_FILE=1")
set(EMSCRIPTEN_LINK_FLAGS "${EMSCRIPTEN_LINK_FLAGS} -s EXPORT_NAME=\"'cv'\" -s DEMANGLE_SUPPORT=1")
)
list(APPEND opencv_test_js_file_deps "${test_data_path}" "${opencv_test_js_bin_dir}/${test_data}")
-add_custom_target(${PROJECT_NAME}_test ALL
+add_custom_target(${PROJECT_NAME}_test
DEPENDS ${OCV_JS_PATH} ${opencv_test_js_file_deps})
# perf
list(APPEND opencv_perf_js_file_deps "${perf_dir}/${f}" "${opencv_perf_js_bin_dir}/${f}")
endforeach()
-add_custom_target(${PROJECT_NAME}_perf ALL
+add_custom_target(${PROJECT_NAME}_perf
DEPENDS ${OCV_JS_PATH} ${opencv_perf_js_file_deps})
#loader
list(APPEND opencv_loader_js_file_deps "${loader_dir}/loader.js" "${opencv_loader_js_bin_dir}/loader.js")
add_custom_target(${PROJECT_NAME}_loader ALL
- DEPENDS ${OCV_JS_PATH} ${opencv_loader_js_file_deps})
\ No newline at end of file
+ DEPENDS ${OCV_JS_PATH} ${opencv_loader_js_file_deps})
+
+add_custom_target(opencv_test_js ALL DEPENDS opencv_js_test opencv_js_perf opencv_js_loader)
--- /dev/null
+# get list of modules to wrap
+if(HAVE_opencv_js)
+ message(STATUS "Wrapped in JavaScript(js):")
+endif()
+set(OPENCV_JS_MODULES "")
+foreach(m ${OPENCV_MODULES_BUILD})
+ if(";${OPENCV_MODULE_${m}_WRAPPERS};" MATCHES ";js;" AND HAVE_${m})
+ list(APPEND OPENCV_JS_MODULES ${m})
+ if(HAVE_opencv_js)
+ message(STATUS " ${m}")
+ endif()
+ endif()
+endforeach()
--- /dev/null
+set(MODULE_NAME "js_bindings_generator")
+set(OPENCV_MODULE_IS_PART_OF_WORLD FALSE)
+ocv_add_module(${MODULE_NAME} INTERNAL)
+
+set(OPENCV_JS_BINDINGS_DIR "${CMAKE_CURRENT_BINARY_DIR}" CACHE INTERNAL "")
+file(REMOVE_RECURSE "${OPENCV_JS_BINDINGS_DIR}/gen")
+file(MAKE_DIRECTORY "${OPENCV_JS_BINDINGS_DIR}/gen")
+file(REMOVE "${OPENCV_DEPHELPER}/gen_opencv_js_source") # force re-run after CMake
+
+# This file is included from a subdirectory
+set(JS_SOURCE_DIR "${CMAKE_CURRENT_SOURCE_DIR}/..")
+include(${JS_SOURCE_DIR}/common.cmake) # fill OPENCV_JS_MODULES
+
+set(opencv_hdrs "")
+foreach(m ${OPENCV_JS_MODULES})
+ list(APPEND opencv_hdrs ${OPENCV_MODULE_${m}_HEADERS})
+endforeach(m)
+
+# header blacklist
+ocv_list_filterout(opencv_hdrs "modules/.*.h$")
+ocv_list_filterout(opencv_hdrs "modules/core/.*/cuda")
+ocv_list_filterout(opencv_hdrs "modules/core/.*/opencl")
+ocv_list_filterout(opencv_hdrs "modules/core/include/opencv2/core/opengl.hpp")
+ocv_list_filterout(opencv_hdrs "modules/core/include/opencv2/core/ocl.hpp")
+ocv_list_filterout(opencv_hdrs "modules/cuda.*")
+ocv_list_filterout(opencv_hdrs "modules/cudev")
+ocv_list_filterout(opencv_hdrs "modules/core/.*/hal/")
+ocv_list_filterout(opencv_hdrs "modules/.*/detection_based_tracker.hpp") # Conditional compilation
+ocv_list_filterout(opencv_hdrs "modules/core/include/opencv2/core/utils/.*")
+
+ocv_update_file("${CMAKE_CURRENT_BINARY_DIR}/headers.txt" "${opencv_hdrs}")
+
+set(bindings_cpp "${OPENCV_JS_BINDINGS_DIR}/gen/bindings.cpp")
+
+set(scripts_hdr_parser "${JS_SOURCE_DIR}/../python/src2/hdr_parser.py")
+
+if(DEFINED ENV{OPENCV_JS_WHITELIST})
+ set(OPENCV_JS_WHITELIST_FILE "$ENV{OPENCV_JS_WHITELIST}")
+else()
+ set(OPENCV_JS_WHITELIST_FILE "${OpenCV_SOURCE_DIR}/platforms/js/opencv_js.config.py")
+endif()
+
+add_custom_command(
+ OUTPUT ${bindings_cpp} "${OPENCV_DEPHELPER}/gen_opencv_js_source"
+ COMMAND
+ ${PYTHON_DEFAULT_EXECUTABLE}
+ "${CMAKE_CURRENT_SOURCE_DIR}/embindgen.py"
+ "${scripts_hdr_parser}"
+ "${bindings_cpp}"
+ "${CMAKE_CURRENT_BINARY_DIR}/headers.txt"
+ "${JS_SOURCE_DIR}/src/core_bindings.cpp"
+ "${OPENCV_JS_WHITELIST_FILE}"
+ COMMAND
+ ${CMAKE_COMMAND} -E touch "${OPENCV_DEPHELPER}/gen_opencv_js_source"
+ WORKING_DIRECTORY
+ "${CMAKE_CURRENT_BINARY_DIR}/gen"
+ DEPENDS
+ ${JS_SOURCE_DIR}/src/core_bindings.cpp
+ ${CMAKE_CURRENT_SOURCE_DIR}/embindgen.py
+ ${CMAKE_CURRENT_SOURCE_DIR}/templates.py
+ ${scripts_hdr_parser}
+ #(not needed - generated by CMake) ${CMAKE_CURRENT_BINARY_DIR}/headers.txt
+ ${opencv_hdrs}
+ COMMENT "Generate source files for JavaScript bindings"
+)
+
+add_custom_target(gen_opencv_js_source
+ # excluded from all: ALL
+ DEPENDS ${bindings_cpp} "${OPENCV_DEPHELPER}/gen_opencv_js_source"
+ SOURCES
+ ${JS_SOURCE_DIR}/src/core_bindings.cpp
+ ${CMAKE_CURRENT_SOURCE_DIR}/embindgen.py
+ ${CMAKE_CURRENT_SOURCE_DIR}/templates.py
+)
return wl
white_list = None
-exec(open(os.environ["OPENCV_JS_WHITELIST"]).read())
-assert(white_list)
# Features to be exported
export_enums = False
if __name__ == "__main__":
- if len(sys.argv) < 4:
+ if len(sys.argv) < 5:
print("Usage:\n", \
os.path.basename(sys.argv[0]), \
- "<full path to hdr_parser.py> <bindings.cpp> <headers.txt> <core_bindings.cpp>")
+ "<full path to hdr_parser.py> <bindings.cpp> <headers.txt> <core_bindings.cpp> <opencv_js.config.py>")
print("Current args are: ", ", ".join(["'"+a+"'" for a in sys.argv]))
exit(0)
bindingsCpp = sys.argv[2]
headers = open(sys.argv[3], 'r').read().split(';')
coreBindings = sys.argv[4]
+ whiteListFile = sys.argv[5]
+ exec(open(whiteListFile).read())
+ assert(white_list)
+
generator = JSWrapperGenerator()
generator.gen(bindingsCpp, headers, coreBindings)
using namespace cv;
#ifdef HAVE_OPENCV_DNN
-using namespace dnn;
+using namespace cv::dnn;
#endif
#ifdef HAVE_OPENCV_ARUCO
# get list of modules to wrap
set(OPENCV_PYTHON_MODULES)
foreach(m ${OPENCV_MODULES_BUILD})
- if (";${OPENCV_MODULE_${m}_WRAPPERS};" MATCHES ";${MODULE_NAME};" AND HAVE_${m})
+ if (";${OPENCV_MODULE_${m}_WRAPPERS};" MATCHES ";python;" AND HAVE_${m})
list(APPEND OPENCV_PYTHON_MODULES ${m})
#message(STATUS "\t${m}")
endif()
def bootstrap():
import sys
+
+ import copy
+ save_sys_path = copy.copy(sys.path)
+
if hasattr(sys, 'OpenCV_LOADER'):
print(sys.path)
raise ImportError('ERROR: recursion is detected during loading of "cv2" binary extensions. Check OpenCV installation.')
del sys.modules['cv2']
import cv2
+ sys.path = save_sys_path # multiprocessing should start from bootstrap code (https://github.com/opencv/opencv/issues/18502)
+
try:
import sys
del sys.OpenCV_LOADER
extern bool skipUnstableTests;
extern bool runBigDataTests;
extern int testThreads;
+extern int debugLevel; //< 0 - no debug, 1 - basic test debug information, >1 - extra debug information
void testSetUp();
void testTearDown();
bool skipUnstableTests = false;
bool runBigDataTests = false;
int testThreads = 0;
+int debugLevel = (int)cv::utils::getConfigurationParameterSizeT("OPENCV_TEST_DEBUG", 0);
static size_t memory_usage_base = 0;
"{ test_threads |-1 |the number of worker threads, if parallel execution is enabled}"
"{ skip_unstable |false |skip unstable tests }"
"{ test_bigdata |false |run BigData tests (>=2Gb) }"
+ "{ test_debug | |0 - no debug (default), 1 - basic test debug information, >1 - extra debug information }"
"{ test_require_data |") + (checkTestData ? "true" : "false") + string("|fail on missing non-required test data instead of skip (env:OPENCV_TEST_REQUIRE_DATA)}"
CV_TEST_TAGS_PARAMS
"{ h help |false |print help info }"
skipUnstableTests = parser.get<bool>("skip_unstable");
runBigDataTests = parser.get<bool>("test_bigdata");
+ if (parser.has("test_debug"))
+ {
+ cv::String s = parser.get<cv::String>("test_debug");
+ if (s.empty() || s == "true")
+ debugLevel = 1;
+ else
+ debugLevel = parser.get<int>("test_debug");
+ }
if (parser.has("test_require_data"))
checkTestData = parser.get<bool>("test_require_data");
}
recordPropertyVerbose("cv_cpu_features", "CPU features", cv::getCPUFeaturesLine());
#ifdef HAVE_IPP
- recordPropertyVerbose("cv_ipp_version", "Intel(R) IPP version", cv::ipp::useIPP() ? cv::ipp::getIppVersion() : "disabled");
+ recordPropertyVerbose("cv_ipp_version", "Intel(R) IPP version", cv::ipp::useIPP() ? cv::ipp::getIppVersion() : "disabled");
+ if (cv::ipp::useIPP())
+ recordPropertyVerbose("cv_ipp_features", "Intel(R) IPP features code", cv::format("0x%llx", cv::ipp::getIppTopFeatures()));
#endif
#ifdef HAVE_OPENCL
cv::dumpOpenCLInformation();
{
case CAP_PROP_POS_FRAMES:
return (double)getFramePos();
+ case CAP_PROP_POS_MSEC:
+ return (double)getFramePos() * (1000. / m_fps);
case CAP_PROP_POS_AVI_RATIO:
return double(getFramePos())/m_mjpeg_frames.size();
case CAP_PROP_FRAME_WIDTH:
STDMETHODIMP OnReadSample(HRESULT hrStatus, DWORD dwStreamIndex, DWORD dwStreamFlags, LONGLONG llTimestamp, IMFSample *pSample) CV_OVERRIDE
{
- CV_UNUSED(llTimestamp);
-
HRESULT hr = 0;
cv::AutoLock lock(m_mutex);
{
CV_LOG_DEBUG(NULL, "videoio(MSMF): drop frame (not processed)");
}
+ m_lastSampleTimestamp = llTimestamp;
m_lastSample = pSample;
}
}
IMFSourceReader *m_reader;
DWORD m_dwStreamIndex;
+ LONGLONG m_lastSampleTimestamp;
_ComPtr<IMFSample> m_lastSample;
};
CV_LOG_WARNING(NULL, "videoio(MSMF): EOS signal. Capture stream is lost");
return false;
}
+ sampleTime = reader->m_lastSampleTimestamp;
return true;
}
else if (isOpen)
namespace opencv_test { namespace {
-static void test_readFrames(/*const*/ VideoCapture& capture, const int N = 100, Mat* lastFrame = NULL)
+static void test_readFrames(/*const*/ VideoCapture& capture, const int N = 100, Mat* lastFrame = NULL, bool testTimestamps = true)
{
Mat frame;
int64 time0 = cv::getTickCount();
+ int64 sysTimePrev = time0;
+ const double cvTickFreq = cv::getTickFrequency();
+
+ double camTimePrev = 0.0;
+ const double fps = capture.get(cv::CAP_PROP_FPS);
+ const double framePeriod = fps == 0.0 ? 1. : 1.0 / fps;
+
+ const bool validTickAndFps = cvTickFreq != 0 && fps != 0.;
+ testTimestamps &= validTickAndFps;
+
for (int i = 0; i < N; i++)
{
SCOPED_TRACE(cv::format("frame=%d", i));
capture >> frame;
+ const int64 sysTimeCurr = cv::getTickCount();
+ const double camTimeCurr = capture.get(cv::CAP_PROP_POS_MSEC);
ASSERT_FALSE(frame.empty());
+ // Do we have a previous frame?
+ if (i > 0 && testTimestamps)
+ {
+ const double sysTimeElapsedSecs = (sysTimeCurr - sysTimePrev) / cvTickFreq;
+ const double camTimeElapsedSecs = (camTimeCurr - camTimePrev) / 1000.;
+
+ // Check that the time between two camera frames and two system time calls
+ // are within 1.5 frame periods of one another.
+ //
+ // 1.5x is chosen to accomodate for a dropped frame, and an additional 50%
+ // to account for drift in the scale of the camera and system time domains.
+ EXPECT_NEAR(sysTimeElapsedSecs, camTimeElapsedSecs, framePeriod * 1.5);
+ }
+
EXPECT_GT(cvtest::norm(frame, NORM_INF), 0) << "Complete black image has been received";
+
+ sysTimePrev = sysTimeCurr;
+ camTimePrev = camTimeCurr;
}
+
int64 time1 = cv::getTickCount();
- printf("Processed %d frames on %.2f FPS\n", N, (N * cv::getTickFrequency()) / (time1 - time0 + 1));
+ printf("Processed %d frames on %.2f FPS\n", N, (N * cvTickFreq) / (time1 - time0 + 1));
if (lastFrame) *lastFrame = frame.clone();
}
else
std::cout << "Frames counter is not available. Actual frames: " << count_actual << ". SKIP check." << std::endl;
}
+
+ void doTimestampTest()
+ {
+ if (!isBackendAvailable(apiPref, cv::videoio_registry::getStreamBackends()))
+ throw SkipTestException(cv::String("Backend is not available/disabled: ") + cv::videoio_registry::getBackendName(apiPref));
+
+ if (((apiPref == CAP_FFMPEG) && ((ext == "h264") || (ext == "h265"))))
+ throw SkipTestException(cv::String("Backend ") + cv::videoio_registry::getBackendName(apiPref) +
+ cv::String(" does not support CAP_PROP_POS_MSEC option"));
+
+ VideoCapture cap;
+ EXPECT_NO_THROW(cap.open(video_file, apiPref));
+ if (!cap.isOpened())
+ throw SkipTestException(cv::String("Backend ") + cv::videoio_registry::getBackendName(apiPref) +
+ cv::String(" can't open the video: ") + video_file);
+
+ Mat img;
+ for(int i = 0; i < 10; i++)
+ {
+ double timestamp = 0;
+ ASSERT_NO_THROW(cap >> img);
+ EXPECT_NO_THROW(timestamp = cap.get(CAP_PROP_POS_MSEC));
+ const double frame_period = 1000.f/bunny_param.getFps();
+ // NOTE: eps == frame_period, because videoCapture returns frame begining timestamp or frame end
+ // timestamp depending on codec and back-end. So the first frame has timestamp 0 or frame_period.
+ EXPECT_NEAR(timestamp, i*frame_period, frame_period);
+ }
+ }
};
//==================================================================================================
TEST_P(videoio_bunny, frame_count) { doFrameCountTest(); }
+TEST_P(videoio_bunny, frame_timestamp) { doTimestampTest(); }
+
INSTANTIATE_TEST_CASE_P(videoio, videoio_bunny,
testing::Combine(
testing::ValuesIn(bunny_params),
"-DBUILD_opencv_superres=OFF",
"-DBUILD_opencv_stitching=OFF",
"-DBUILD_opencv_java=OFF",
- "-DBUILD_opencv_java_bindings_generator=OFF",
"-DBUILD_opencv_js=ON",
"-DBUILD_opencv_python2=OFF",
"-DBUILD_opencv_python3=OFF",
- "-DBUILD_opencv_python_bindings_generator=OFF",
"-DBUILD_EXAMPLES=OFF",
"-DBUILD_PACKAGE=OFF",
"-DBUILD_TESTS=OFF",
projects / platform / upstream / opencv.git / commitdiff