generic-library/opencv
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

optimizations:

Browse Source 
- new reduce implementation (with kepler optimizations)
- saturate_cast via asm command
- video SIMD instructions in element operations
- float arithmetics instead of double
- new deviceSupports function
This commit is contained in:
Vladislav Vinogradov
2013-01-23 14:43:36 +04:00
parent ae6266e101
commit 281d036fcf
45 changed files with 9379 additions and 8846 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
modules/gpu/src/remap.cpp
View File

@@ -54,7 +54,7 @@ namespace cv { namespace gpu { namespace device
{
template <typename T>
void remap_gpu(PtrStepSzb src, PtrStepSzb srcWhole, int xoff, int yoff, PtrStepSzf xmap, PtrStepSzf ymap, PtrStepSzb dst,
int interpolation, int borderMode, const float* borderValue, cudaStream_t stream, int cc);
int interpolation, int borderMode, const float* borderValue, cudaStream_t stream, bool cc20);
}
}}}
@@ -63,7 +63,7 @@ void cv::gpu::remap(const GpuMat& src, GpuMat& dst, const GpuMat& xmap, const Gp
using namespace cv::gpu::device::imgproc;
typedef void (*func_t)(PtrStepSzb src, PtrStepSzb srcWhole, int xoff, int yoff, PtrStepSzf xmap, PtrStepSzf ymap, PtrStepSzb dst, int interpolation,
int borderMode, const float* borderValue, cudaStream_t stream, int cc);
int borderMode, const float* borderValue, cudaStream_t stream, bool cc20);
static const func_t funcs[6][4] =
{
@@ -91,15 +91,12 @@ void cv::gpu::remap(const GpuMat& src, GpuMat& dst, const GpuMat& xmap, const Gp
Scalar_<float> borderValueFloat;
borderValueFloat = borderValue;
DeviceInfo info;
int cc = info.majorVersion() * 10 + info.minorVersion();
Size wholeSize;
Point ofs;
src.locateROI(wholeSize, ofs);
func(src, PtrStepSzb(wholeSize.height, wholeSize.width, src.datastart, src.step), ofs.x, ofs.y, xmap, ymap,
dst, interpolation, gpuBorderType, borderValueFloat.val, StreamAccessor::getStream(stream), cc);
dst, interpolation, gpuBorderType, borderValueFloat.val, StreamAccessor::getStream(stream), deviceSupports(FEATURE_SET_COMPUTE_20));
}
#endif // HAVE_CUDA