From e4cfabafcdde83b94ae5002ace0f178694eca2e9 Mon Sep 17 00:00:00 2001 From: "jy1210.jung" Date: Tue, 21 Aug 2018 17:30:32 +0900 Subject: [PATCH] [Doc] add link to examples wiki page [how-to-run-examples.md] link to examples wiki page (nnsuite/nnstreamer) **Self evaluation:** 1. Build test: [ ]Passed [ ]Failed [*]Skipped 2. Run test: [ ]Passed [ ]Failed [*]Skipped Signed-off-by: Jaeyun Jung --- Documentation/how-to-run-examples.md | 115 +---------------------------------- 1 file changed, 1 insertion(+), 114 deletions(-) diff --git a/Documentation/how-to-run-examples.md b/Documentation/how-to-run-examples.md index 784361e..fed444e 100644 --- a/Documentation/how-to-run-examples.md +++ b/Documentation/how-to-run-examples.md @@ -1,116 +1,3 @@ # How to run examples -## Build examples (Ubuntu) - -Use Cmake. - -See [Getting started](getting-started.md) : **Approach 2.** Build with Cmake - -- Build example (set your path for NNST_ROOT, then gst plugin path is ```$NNST_ROOT/lib```) - -``` -# prepare -$ sudo apt-get install python-gi python3-gi # for python example -$ sudo apt-get install python-gst-1.0 python3-gst-1.0 -$ sudo apt-get install python-gst-1.0-dbg python3-gst-1.0-dbg -$ export NNST_ROOT=$HOME/nnstreamer # set your own path -$ export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$NNST_ROOT/lib -$ export GST_PLUGIN_PATH=$GST_PLUGIN_PATH:$NNST_ROOT/lib -``` -``` -# build example -$ cd nnstreamer -$ mkdir -p build # We recommend to build in a "build" directory -$ cd build -$ rm -rf * # Ensure the build directory is empty -$ cmake -DCMAKE_INSTALL_PREFIX=${NNST_ROOT} -DINCLUDE_INSTALL_DIR=${NNST_ROOT}/include .. -$ make install # Install nnstreamer plugin libraries into $NNST_ROOT/lib -$ cd .. -``` - -## Example : filter for image classification - -``` -v4l2src -- tee -- textoverlay -- videoconvert -- ximagesink - | - --- videoscale -- tensor_converter -- tensor_filter -- tensor_sink -``` - -NNStreamer example for image recognition. -- Download tflite moel 'Mobilenet_1.0_224_quant' from [Here](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/lite/g3doc/models.md#image-classification-quantized-models) - -Displays video sink. - -1. 'tensor_filter' for image recognition. (classification with 224x224 image). -2. 'tensor_sink' updates recognition result to display in textoverlay. - -- Run example -``` -$ cd build/nnstreamer_example/example_filter -$ ./nnstreamer_example_filter -``` - -``` -# for python example -$ cd nnstreamer_example/example_filter -$ python nnstreamer_example_filter.py -``` - -## Example : video mixer with NNStreamer plug-in - -``` -v4l2src -- tee ------------------------------------------ videomixer -- videoconvert -- ximagesink (Mixed) - | | - --- tensor_converter -- tensordec -- videoscale --- - | - --- videoconvert -- ximagesink (Original) -``` - -Displays two video sinks, - -1. Original from cam -2. Mixed : original + scaled (tensor_converter-tensor_dec-videoscale) - -In pipeline, converter-decoder passes video frame. - -- Run example -``` -$ cd build/nnstreamer_example/example_cam -$ ./nnstreamer_example_cam -``` - -## Example : tensor sink - -Two simple examples to use tensor sink. - -#### 1. 100 buffers passed to tensor sink - -``` -videotestsrc -- tensor_converter -- tensor_sink -``` - -Displays nothing, this sample code shows how to get buffer from tensor sink. - -- Run example -``` -$ cd build/nnstreamer_example/example_sink -$ ./nnstreamer_sink_example -``` - -#### 2. launch two pipelines - -``` -videotestsrc -- tensor_converter -- tensor_sink -[push buffer from tensor_sink to appsrc] -appsrc -- tensordec -- videoconvert -- ximagesink -``` - -Displays video sink. - -Tensor sink receives buffer and pushes it into appsrc in 2nd pipeline. - -- Run example -``` -$ cd build/nnstreamer_example/example_sink -$ ./nnstreamer_sink_example_play -``` +See nnstreamer wiki page [Usage examples and screenshots](https://github.com/nnsuite/nnstreamer/wiki/usage-examples-screenshots) -- 2.7.4