search

Bkmz21 / CompactCNNCascade

A binary library for very fast face detection using compact CNNs.
Other
156 stars 50 forks source link

Segmentation fault (core dumped) #8

Open dexception opened 5 years ago

dexception commented 5 years ago

I already have opencv 3.4.0 installed

root@xyz:/home/xyz# pkg-config --cflags opencv -I/usr/local/include/opencv -I/usr/local/include

root@xyz:/home/xyz# pkg-config --libs opencv -L/usr/local/lib -lopencv_stitching -lopencv_superres -lopencv_videostab -lopencv_aruco -lopencv_bgsegm -lopencv_bioinspired -lopencv_ccalib -lopencv_dpm -lopencv_face -lopencv_photo -lopencv_freetype -lopencv_fuzzy -lopencv_hdf -lopencv_img_hash -lopencv_line_descriptor -lopencv_optflow -lopencv_reg -lopencv_rgbd -lopencv_saliency -lopencv_stereo -lopencv_structured_light -lopencv_phase_unwrapping -lopencv_surface_matching -lopencv_tracking -lopencv_datasets -lopencv_text -lopencv_dnn -lopencv_plot -lopencv_xfeatures2d -lopencv_shape -lopencv_video -lopencv_ml -lopencv_ximgproc -lopencv_calib3d -lopencv_features2d -lopencv_highgui -lopencv_videoio -lopencv_flann -lopencv_xobjdetect -lopencv_imgcodecs -lopencv_objdetect -lopencv_xphoto -lopencv_imgproc -lopencv_core

git clone https://github.com/Bkmz21/CompactCNNCascade cd CompactCNNCascade mkdir build cd build cmake .. -- The C compiler identification is GNU 7.3.0 -- The CXX compiler identification is GNU 7.3.0 -- Check for working C compiler: /usr/bin/cc -- Check for working C compiler: /usr/bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Check for working CXX compiler: /usr/bin/c++ -- Check for working CXX compiler: /usr/bin/c++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Configuring done -- Generating done -- Build files have been written to: /home/xyz/Downloads/github/CompactCNNCascade/build root@xyz:/home/xyz# make LOTS OF ERRORS

n file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:44:138: error: ‘size_t’ has not been declared void conv_3x3(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:44:152: error: ‘size_t’ has not been declared void conv_3x3(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:45:138: error: ‘size_t’ has not been declared void conv_4x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:45:152: error: ‘size_t’ has not been declared void conv_4x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:46:138: error: ‘size_t’ has not been declared void conv_5x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:46:152: error: ‘size_t’ has not been declared void conv_5x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:47:138: error: ‘size_t’ has not been declared void conv_5x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:47:152: error: ‘size_t’ has not been declared void conv_5x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:48:138: error: ‘size_t’ has not been declared void conv_6x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:48:152: error: ‘size_t’ has not been declared void conv_6x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:49:138: error: ‘size_t’ has not been declared void conv_6x6(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:49:152: error: ‘size_t’ has not been declared void conv_6x6(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:50:138: error: ‘size_t’ has not been declared void conv_7x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:50:152: error: ‘size_t’ has not been declared void conv_7x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:51:138: error: ‘size_t’ has not been declared void conv_8x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:51:152: error: ‘size_t’ has not been declared void conv_8x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:52:138: error: ‘size_t’ has not been declared void conv_8x8(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:52:152: error: ‘size_t’ has not been declared void conv_8x8(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:53:140: error: ‘size_t’ has not been declared void conv_11x10(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:53:154: error: ‘size_t’ has not been declared void conv_11x10(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:54:140: error: ‘size_t’ has not been declared void conv_11x11(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:54:154: error: ‘size_t’ has not been declared void conv_11x11(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:36:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_3x3(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_3x3(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:44:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_3x3(float*, int, float*, int, int, float*, int, int) void conv_3x3(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:92:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_4x4(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_4x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:45:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_4x4(float*, int, float*, int, int, float*, int, int) void conv_4x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:158:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_5x4(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_5x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:46:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_5x4(float*, int, float*, int, int, float*, int, int) void conv_5x4(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:188:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_5x5(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_5x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:47:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_5x5(float*, int, float*, int, int, float*, int, int) void conv_5x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:212:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_6x5(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_6x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:48:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_6x5(float*, int, float*, int, int, float*, int, int) void conv_6x5(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:238:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_6x6(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_6x6(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:49:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_6x6(float*, int, float*, int, int, float*, int, int) void conv_6x6(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:264:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_7x7(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_7x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:50:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_7x7(float*, int, float*, int, int, float*, int, int) void conv_7x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:292:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_8x7(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_8x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:51:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_8x7(float*, int, float*, int, int, float*, int, int) void conv_8x7(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:322:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_8x8(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_8x8(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:52:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_8x8(float*, int, float*, int, int, float*, int, int) void conv_8x8(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:352:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_11x10(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_11x10(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:53:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_11x10(float*, int, float*, int, int, float*, int, int) void conv_11x10(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~~~ /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:381:8: error: prototype for ‘void NeuralNetworksLib::SIMD::CNNPP::conv_11x11(float*, int, float*, int, int, float*, size_t, size_t)’ does not match any in class ‘NeuralNetworksLib::SIMD::CNNPP’ void CNNPP::conv_11x11(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L, size_t H) ^~~~~ In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp:21:0: /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h:54:9: error: candidate is: void NeuralNetworksLib::SIMD::CNNPP::conv_11x11(float*, int, float*, int, int, float*, int, int) void conv_11x11(float* __restrict dst, int dst_size_l, float* __restrict src, int src_size_l, int src_size_h, float* __restrict kernel, size_t L = 0, size_t H = 0); ^~~~~~~~~~ CMakeFiles/CNNObjectDetector.dir/build.make:278: recipe for target 'CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/cnnpp_cplusplus.cpp.o' failed make[2]: *** [CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/cnnpp_cplusplus.cpp.o] Error 1 CMakeFiles/Makefile2:67: recipe for target 'CMakeFiles/CNNObjectDetector.dir/all' failed make[1]: *** [CMakeFiles/CNNObjectDetector.dir/all] Error 2 Makefile:83: recipe for target 'all' failed make: *** [all] Error 2

Edited vim github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.cpp vim github/CompactCNNCascade/src/CNNObjectDetector/cnnpp_cplusplus.h Add #include for size_t error

root@xyz:/home/xyz# make LOTS OF ERRORS

Scanning dependencies of target CNNObjectDetector
[  5%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/test_accuracy.cpp.o
In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/test_accuracy.cpp:75:0:
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/cnn_detector_v3.h:387:15: warning: inline function ‘void NeuralNetworksLib::CNNDetector::Merger(std::vector<NeuralNetworksLib::CNNDetector::Detection>&, std::vector<NeuralNetworksLib::CNNDetector::Detection>&, float, bool)’ used but never defined
   inline void Merger(std::vector<Detection>& detections, std::vector<Detection>& rect, float threshold = 0.5f, bool del = false);
               ^~~~~~
[ 10%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/cnn_simd.cpp.o
[ 15%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/cnn_simd_cntk.cpp.o
[ 21%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/cnnpp_cplusplus.cpp.o
[ 26%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/image.cpp.o
[ 31%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/image_proc.cpp.o
[ 36%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/image_proc_cl.cpp.o
[ 42%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/image_resize.cpp.o
[ 47%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/image_resize_cl.cpp.o
[ 52%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/timer.cpp.o
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp: In member function ‘void NeuralNetworksLib::Timer::start(int)’:
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:50:69: error: no match for ‘operator=’ (operand types are ‘std::chrono::_V2::steady_clock::time_point {aka std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >}’ and ‘std::chrono::_V2::system_clock::time_point {aka std::chrono::time_point<std::chrono::_V2::system_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >}’)
   this->counters[counter] = std::chrono::high_resolution_clock::now();
                                                                     ^
In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.h:33:0,
                 from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:21:
/usr/include/c++/7/chrono:616:14: note: candidate: std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >& std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >::operator=(const std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >&)
       struct time_point
              ^~~~~~~~~~
/usr/include/c++/7/chrono:616:14: note:   no known conversion for argument 1 from ‘std::chrono::_V2::system_clock::time_point {aka std::chrono::time_point<std::chrono::_V2::system_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >}’ to ‘const std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >&’
/usr/include/c++/7/chrono:616:14: note: candidate: std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >& std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >::operator=(std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >&&)
/usr/include/c++/7/chrono:616:14: note:   no known conversion for argument 1 from ‘std::chrono::_V2::system_clock::time_point {aka std::chrono::time_point<std::chrono::_V2::system_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >}’ to ‘std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >&&’
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp: In member function ‘double NeuralNetworksLib::Timer::get(double, int)’:
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:55:44: error: no match for ‘operator-’ (operand types are ‘std::chrono::time_point<std::chrono::_V2::system_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >’ and ‘std::chrono::_V2::steady_clock::time_point {aka std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >}’)
   std::chrono::duration<double> diff = end - this->counters[counter];
                                        ~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.h:33:0,
                 from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:21:
/usr/include/c++/7/chrono:752:7: note: candidate: template<class _Clock, class _Dur1, class _Dur2> constexpr typename std::common_type<_Duration1, _Duration2>::type std::chrono::operator-(const std::chrono::time_point<_Clock, _Duration1>&, const std::chrono::time_point<_Clock, _Duration2>&)
       operator-(const time_point<_Clock, _Dur1>& __lhs,
       ^~~~~~~~
/usr/include/c++/7/chrono:752:7: note:   template argument deduction/substitution failed:
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:55:68: note:   deduced conflicting types for parameter ‘_Clock’ (‘std::chrono::_V2::system_clock’ and ‘std::chrono::_V2::steady_clock’)
   std::chrono::duration<double> diff = end - this->counters[counter];
                                                                    ^
In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.h:33:0,
                 from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:21:
/usr/include/c++/7/chrono:741:7: note: candidate: template<class _Clock, class _Dur1, class _Rep2, class _Period2> constexpr std::chrono::time_point<_Clock, typename std::common_type<_Dur1, std::chrono::duration<_Rep2, _Period2> >::type> std::chrono::operator-(const std::chrono::time_point<_Clock, _Duration1>&, const std::chrono::duration<_Rep2, _Period2>&)
       operator-(const time_point<_Clock, _Dur1>& __lhs,
       ^~~~~~~~
/usr/include/c++/7/chrono:741:7: note:   template argument deduction/substitution failed:
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:55:68: note:   ‘std::chrono::_V2::steady_clock::time_point {aka std::chrono::time_point<std::chrono::_V2::steady_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >}’ is not derived from ‘const std::chrono::duration<_Rep2, _Period2>’
   std::chrono::duration<double> diff = end - this->counters[counter];
                                                                    ^
In file included from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.h:33:0,
                 from /home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:21:
/usr/include/c++/7/chrono:459:7: note: candidate: template<class _Rep1, class _Period1, class _Rep2, class _Period2> constexpr typename std::common_type<std::chrono::duration<_Rep1, _Period1>, std::chrono::duration<_Rep2, _Period2> >::type std::chrono::operator-(const std::chrono::duration<_Rep1, _Period1>&, const std::chrono::duration<_Rep2, _Period2>&)
       operator-(const duration<_Rep1, _Period1>& __lhs,
       ^~~~~~~~
/usr/include/c++/7/chrono:459:7: note:   template argument deduction/substitution failed:
/home/xyz/Downloads/github/CompactCNNCascade/src/CNNObjectDetector/timer.cpp:55:68: note:   ‘std::chrono::time_point<std::chrono::_V2::system_clock, std::chrono::duration<long int, std::ratio<1, 1000000000> > >’ is not derived from ‘const std::chrono::duration<_Rep1, _Period1>’
   std::chrono::duration<double> diff = end - this->counters[counter];
                                                                    ^
CMakeFiles/CNNObjectDetector.dir/build.make:422: recipe for target 'CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/timer.cpp.o' failed
make[2]: *** [CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/timer.cpp.o] Error 1
CMakeFiles/Makefile2:67: recipe for target 'CMakeFiles/CNNObjectDetector.dir/all' failed
make[1]: *** [CMakeFiles/CNNObjectDetector.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2

Edited the following files vim CompactCNNCascade/src/CNNObjectDetector/timer.cpp Replaced high_resolution_clock with steady_clock

All Code compiled perfectly.

root@xyz:/home/xyz# make Scanning dependencies of target CNNObjectDetector [ 5%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/timer.cpp.o [ 10%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/cnn_detector_v3.cpp.o [ 15%] Building CXX object CMakeFiles/CNNObjectDetector.dir/src/CNNObjectDetector/packing_2D.cpp.o [ 21%] Linking CXX executable bin/CNNObjectDetector [100%] Built target CNNObjectDetector

root@xyz:/home/xyz# cd bin

root@xyz:/home/xyz# ./CNNObjectDetector 
TEST ACCURACY

[CNNDetector] Initializing with max image size (1280, 720)!

[TEST ACCURACY]  test allocate memory
[TEST ACCURACY]     cannels = 1: success
[TEST ACCURACY]     cannels = 3: success
[TEST ACCURACY]     cannels = 4: success

[TEST ACCURACY]  test image converter
Segmentation fault (core dumped)
jamessmith90 commented 5 years ago

+1 Same Issue

@Bkmz21 Any resolutions ?

Bkmz21 commented 5 years ago

Hi, I can not catch the error on my system. Can you tell which function the error occurs in?

dexception commented 5 years ago

@Bkmz21 Thanks for replying !

root@xyz:/home/xyz# ./CNNObjectDetector 
TEST ACCURACY

[CNNDetector] Initializing with max image size (1280, 720)!

[TEST ACCURACY]  test allocate memory
[TEST ACCURACY]     cannels = 1: success
[TEST ACCURACY]     cannels = 3: success
[TEST ACCURACY]     cannels = 4: success

[TEST ACCURACY]  test image converter
Segmentation fault (core dumped)

https://github.com/Bkmz21/CompactCNNCascade/blob/master/src/CNNObjectDetector/test_accuracy.cpp#L350

Bkmz21 commented 5 years ago

Surely one of these functions falls. https://github.com/Bkmz21/CompactCNNCascade/blob/master/src/CNNObjectDetector/test_accuracy.cpp#L405-L437 You can comment them in sequence until this block of checks is passed.

dexception commented 5 years ago

@Bkmz21 Disabled Image Converter, Resizer, CNN [All of them are giving segmentation fault]

Changed BUILD_FaceDetector to ON inside CMakeLists.txt but there is not binary to test it.

Please review the code.. It is not stable at all.

Bkmz21 commented 5 years ago

Thanks, I'll try to fix it in a week.

dexception commented 5 years ago

@Bkmz21 Tried it on windows. Not accurate as claimed in the paper.

Bkmz21 commented 5 years ago

Hi,

You must change the scale_factor and minNeighbour parameters to improve accuracy.

dexception commented 5 years ago

@Bkmz21

My Observations:

Ran on a Camera: Size max_image_size = Size(2560, 1440);

*******************************************************************
* Compact Convolutional Neural Network Cascade for Face Detection *
* http://arxiv.org/abs/1508.01292                                 *
* Copyright (c) 2017, Ilya Kalinovskiy, e-mail: kua_21@mail.ru    *
* All rights reserved                                             *
*******************************************************************

[CompactCNNLibAPI] CNN Cascade v2.0
[CompactCNNLibAPI] Face Detector v3.1
[CompactCNNLibAPI] API v2.0
[CompactCNNLibAPI] SIMD support (AVX)!
[CompactCNNLibAPI] CUDA support!
[CompactCNNLibAPI] OpenMP support (maximum 4 threads)!

[CNNDetector] Initializing with max image size (2560, 1440)!
[CNNDetector] [CUDA] Device count: 1

[CNNDetector] [CUDA] Device id: 0
[CNNDetector] [CUDA] Device name: GeForce 940MX
[CNNDetector] [CUDA] Total global memory: -2147483648
[CNNDetector] [CUDA] Shared memory per block: 49152
[CNNDetector] [CUDA] Registers per block: 65536
[CNNDetector] [CUDA] Warp size: 32
[CNNDetector] [CUDA] Memory pitch: 2147483647
[CNNDetector] [CUDA] Max threads per block: 1024
[CNNDetector] [CUDA] Max threads dimensions: x = 1024, y = 1024, z = 64
[CNNDetector] [CUDA] Max grid size: x = 2147483647, y = 65535, z = 65535
[CNNDetector] [CUDA] Clock rate: 1189000
[CNNDetector] [CUDA] Total constant memory: 65536
[CNNDetector] [CUDA] Compute capability: 5.0
[CNNDetector] [CUDA] Texture alignment: 512
[CNNDetector] [CUDA] Device overlap: 1
[CNNDetector] [CUDA] Multiprocessor count: 3
[CNNDetector] [CUDA] Kernel execution concurrent support: true
[CNNDetector] [CUDA] Kernel execution timeout enabled: true

[CNNDetector] [CUDA] CUDA device id selected: 0
time detect: 42.623 ms
time detect: 41.631 ms
time detect: 41.831 ms

With the following settings:

    param.min_face_height = 50;
    param.scale_factor = 1.1f;
    param.equalize = false;
    param.reflection = false;
    param.pipeline = CompactCNNLib::FaceDetector::Pipeline::GPU;
    param.num_threads = 4;
    param.detect_precision = CompactCNNLib::FaceDetector::DetectPrecision::high;
    param.facial_analysis = false;
  1. Got lots of false positives.
  2. Missing front faces with beard also Side faces are missed.

Am i using the wrong settings ?

dexception commented 5 years ago

@Bkmz21 Any plan to fix the bugs ?