1 /*M///////////////////////////////////////////////////////////////////////////////////////
\r
3 // IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
\r
5 // By downloading, copying, installing or using the software you agree to this license.
\r
6 // If you do not agree to this license, do not download, install,
\r
7 // copy or use the software.
\r
10 // License Agreement
\r
11 // For Open Source Computer Vision Library
\r
13 // Copyright (C) 2000-2008, Intel Corporation, all rights reserved.
\r
14 // Copyright (C) 2009, Willow Garage Inc., all rights reserved.
\r
15 // Third party copyrights are property of their respective owners.
\r
17 // Redistribution and use in source and binary forms, with or without modification,
\r
18 // are permitted provided that the following conditions are met:
\r
20 // * Redistribution's of source code must retain the above copyright notice,
\r
21 // this list of conditions and the following disclaimer.
\r
23 // * Redistribution's in binary form must reproduce the above copyright notice,
\r
24 // this list of conditions and the following disclaimer in the documentation
\r
25 // and/or other materials provided with the distribution.
\r
27 // * The name of the copyright holders may not be used to endorse or promote products
\r
28 // derived from this software without specific prior written permission.
\r
30 // This software is provided by the copyright holders and contributors "as is" and
\r
31 // any express or bpied warranties, including, but not limited to, the bpied
\r
32 // warranties of merchantability and fitness for a particular purpose are disclaimed.
\r
33 // In no event shall the Intel Corporation or contributors be liable for any direct,
\r
34 // indirect, incidental, special, exemplary, or consequential damages
\r
35 // (including, but not limited to, procurement of substitute goods or services;
\r
36 // loss of use, data, or profits; or business interruption) however caused
\r
37 // and on any theory of liability, whether in contract, strict liability,
\r
38 // or tort (including negligence or otherwise) arising in any way out of
\r
39 // the use of this software, even if advised of the possibility of such damage.
\r
43 #include "internal_shared.hpp"
\r
45 using namespace cv::gpu;
\r
47 namespace cv { namespace gpu
\r
50 template <typename T>
\r
51 __global__ void blendLinearKernel(int rows, int cols, int cn, const PtrStep_<T> img1, const PtrStep_<T> img2,
\r
52 const PtrStepf weights1, const PtrStepf weights2, PtrStep_<T> result)
\r
54 int x = blockIdx.x * blockDim.x + threadIdx.x;
\r
55 int y = blockIdx.y * blockDim.y + threadIdx.y;
\r
57 if (y < rows && x < cols)
\r
60 float w1 = weights1.ptr(y)[x_];
\r
61 float w2 = weights2.ptr(y)[x_];
\r
62 T p1 = img1.ptr(y)[x];
\r
63 T p2 = img2.ptr(y)[x];
\r
64 result.ptr(y)[x] = (p1 * w1 + p2 * w2) / (w1 + w2 + 1e-5f);
\r
69 template <typename T>
\r
70 void blendLinearCaller(int rows, int cols, int cn, const PtrStep_<T> img1, const PtrStep_<T> img2,
\r
71 const PtrStepf weights1, const PtrStepf weights2, PtrStep_<T> result, cudaStream_t stream)
\r
73 dim3 threads(16, 16);
\r
74 dim3 grid(divUp(cols * cn, threads.x), divUp(rows, threads.y));
\r
76 blendLinearKernel<<<grid, threads, 0, stream>>>(rows, cols * cn, cn, img1, img2, weights1, weights2, result);
\r
77 cudaSafeCall( cudaGetLastError() );
\r
80 cudaSafeCall(cudaDeviceSynchronize());
\r
83 template void blendLinearCaller<uchar>(int, int, int, const PtrStep, const PtrStep,
\r
84 const PtrStepf, const PtrStepf, PtrStep, cudaStream_t stream);
\r
85 template void blendLinearCaller<float>(int, int, int, const PtrStepf, const PtrStepf,
\r
86 const PtrStepf, const PtrStepf, PtrStepf, cudaStream_t stream);
\r
89 __global__ void blendLinearKernel8UC4(int rows, int cols, const PtrStep img1, const PtrStep img2,
\r
90 const PtrStepf weights1, const PtrStepf weights2, PtrStep result)
\r
92 int x = blockIdx.x * blockDim.x + threadIdx.x;
\r
93 int y = blockIdx.y * blockDim.y + threadIdx.y;
\r
95 if (y < rows && x < cols)
\r
97 float w1 = weights1.ptr(y)[x];
\r
98 float w2 = weights2.ptr(y)[x];
\r
99 float sum_inv = 1.f / (w1 + w2 + 1e-5f);
\r
102 uchar4 p1 = ((const uchar4*)img1.ptr(y))[x];
\r
103 uchar4 p2 = ((const uchar4*)img2.ptr(y))[x];
\r
104 ((uchar4*)result.ptr(y))[x] = make_uchar4(p1.x * w1 + p2.x * w2, p1.y * w1 + p2.y * w2,
\r
105 p1.z * w1 + p2.z * w2, p1.w * w1 + p2.w * w2);
\r
110 void blendLinearCaller8UC4(int rows, int cols, const PtrStep img1, const PtrStep img2,
\r
111 const PtrStepf weights1, const PtrStepf weights2, PtrStep result, cudaStream_t stream)
\r
113 dim3 threads(16, 16);
\r
114 dim3 grid(divUp(cols, threads.x), divUp(rows, threads.y));
\r
116 blendLinearKernel8UC4<<<grid, threads, 0, stream>>>(rows, cols, img1, img2, weights1, weights2, result);
\r
117 cudaSafeCall( cudaGetLastError() );
\r
120 cudaSafeCall(cudaDeviceSynchronize());
\r