【视频开发】【计算机视觉】doppia编译之四:安装其他库、编译和运行doppia
2017-06-28 01:17
302 查看
(与本节内容无关///////////////////////////保存图片参数为—-gui.save_all_screenshots true////////////////////)
在我们安装好CUDA、boost、OpenCV之后,接下来的一些库(libSDL、protobuf等)的安装,我们都可以用系统内部的程序进行安装。比如
安装libSDL,我们终端输入
1
系统会给出一系列程序,我们选择其中的libsdl1.2-dev进行安装。
1
(这里注意不要安装libsdl2-dev,因为安装之后生成的文件夹是SDL2,之后doppia调用时会出现找不到“SDL files”的错误)
安装protobuf库,则是我经过多番测试,得到的能够通过v1,v2测试的安装方法。(之前尝试安装protobuf2.5.0和protobuf2.4.1,doppia都找不到路径,而想把它们删除又删除不了,很麻烦),之后我测试了几个自带的protobuf库,发现安装以下四个库能够通过v1,v2的测试,安装命令为:
1
切换到doppia目录下,运行
1
protobuf通过doppia-v1检测的返回信息为
2
3
1
2
3
protobuf通过doppia-v2检测的返回信息为
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
到这里,我们该安装的库大部分已经安装成功,接下来就可以开始编译doppia啦!(至于可能还缺少的库,可以根据doppia的错误提示进行安装)
编译运行doppia/src/applications/objects_detection
由于我只需要用到doppia的objects_detection的功能,而之前我在编译doppia-v2时,ground_estimation和stixel_world都能编译运行。所以这次在编译doppia-v1时,我就直接切入“主题”,编译运行objects_detection。下面也主要是列出我在编译objects_detection是遇到的问题以及相应的解决方案。
错误一,创建(build)错误
error:
2
1
2
solution:
在
1
头文件中,在
2
1
2
两行之间添加一行新的引用,如下,
2
3
1
2
3
错误二,创建(build)错误
error:
2
3
4
5
6
7
8
9
10
11
1
2
3
4
5
6
7
8
9
10
11
这里出现错误原因是因为common_settings.cmake中没有添加cuda链接库路径。
solution:
编辑common_settings.cmake,在其中添加一个条件项
2
3
4
5
6
7
1
2
3
4
5
6
7
这里mx-pc是我电脑的主机名,你需要将它改成自己电脑的主机名。
错误三,创建(build)错误
error:
2
1
2
solution:
在
1
文件开头添加一行引用
1
解决了以上三个错误后,doppia就可以创建(build)成功啦!但要想运行成功,还得改正以下2个错误。
错误四,链接(link)错误
error:
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
这里出现错误的原因,是boost库链接出错,这时候我们需要修改CMakeList.txt文件,这里我就直接把CMakeList.txt贴出来,修改的地方做过注释。
solution:
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
到这里就OK啦!
最后运行成功你会看到一个简短地video,以及下面这样的信息
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
好啦,doppia编译到此也就结束啦!
希望这几篇文章能帮助正在读博客的你。
doppia及作者相关介绍链接:
http://blog.csdn.net/xizero00/article/details/43227019
https://bitbucket.org/rodrigob/doppia
在我们安装好CUDA、boost、OpenCV之后,接下来的一些库(libSDL、protobuf等)的安装,我们都可以用系统内部的程序进行安装。比如
安装libSDL,我们终端输入
apt-cache search libsdl1
1
系统会给出一系列程序,我们选择其中的libsdl1.2-dev进行安装。
sudo apt-get install libsdl1.2-dev1
1
(这里注意不要安装libsdl2-dev,因为安装之后生成的文件夹是SDL2,之后doppia调用时会出现找不到“SDL files”的错误)
安装protobuf库,则是我经过多番测试,得到的能够通过v1,v2测试的安装方法。(之前尝试安装protobuf2.5.0和protobuf2.4.1,doppia都找不到路径,而想把它们删除又删除不了,很麻烦),之后我测试了几个自带的protobuf库,发现安装以下四个库能够通过v1,v2的测试,安装命令为:
sudo apt-get install libprotobuf-dev libprotoc-dev python-protobuf protobuf-compiler1
1
切换到doppia目录下,运行
sudo sh ./generate_protocol_buffer_files.sh1
1
protobuf通过doppia-v1检测的返回信息为
Generating objects detection files... (Ground plane and video input files not yet handled by this script) End of game. Have a nice day!1
2
3
1
2
3
protobuf通过doppia-v2检测的返回信息为
+ cd src/objects_detection/ + protoc --cpp_out=./ detector_model.proto detections.proto + protoc --python_out=../../tools/objects_detection/ detector_model.proto detections.proto + cd ../.. + cd src/stereo_matching/ground_plane/ + protoc --cpp_out=./ plane3d.proto + protoc --python_out=../../../tools/stixels_evaluation plane3d.proto + cd ../../.. + cd src/stereo_matching/stixels/ + protoc --cpp_out=./ -I. -I../ground_plane --include_imports stixels.proto ground_top_and_bottom.proto --include_imports only makes sense when combined with --descriptor_set_out. + protoc --python_out=../../../tools/stixels_evaluation -I. -I../ground_plane --include_imports stixels.proto ground_top_and_bottom.proto --include_imports only makes sense when combined with --descriptor_set_out. + cd ../../.. + cd src/video_input/calibration + protoc --cpp_out=./ calibration.proto + cd ../../.. + cd src/helpers/data + protoc --cpp_out=./ DataSequenceHeader.proto + protoc --python_out=../../../tools/data_sequence DataSequenceHeader.proto + cd ../../.. + cd src/helpers + cd ../.. + cd src/tests/data_sequence/ + protoc --cpp_out=./ TestData.proto + cd ../../.. + echo End of game. Have a nice day! End of game. Have a nice day!1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
到这里,我们该安装的库大部分已经安装成功,接下来就可以开始编译doppia啦!(至于可能还缺少的库,可以根据doppia的错误提示进行安装)
编译运行doppia/src/applications/objects_detection
由于我只需要用到doppia的objects_detection的功能,而之前我在编译doppia-v2时,ground_estimation和stixel_world都能编译运行。所以这次在编译doppia-v1时,我就直接切入“主题”,编译运行objects_detection。下面也主要是列出我在编译objects_detection是遇到的问题以及相应的解决方案。
错误一,创建(build)错误
error:
/home/mx/doppia/src/applications/objects_detection/../../../src/helpers/data/DataSequence.hpp:293:56: error: invalid use of incomplete type ‘class google::protobuf::io::CodedInputStream’ const bool read_size_success = input_coded_stream_p->ReadLittleEndian64(&size);1
2
1
2
solution:
在
doppia/src/helpers/data/DataSequence.hpp1
1
头文件中,在
#include "DataSequenceHeader.pb.h" #include <google/protobuf/io/zero_copy_stream_impl.h>1
2
1
2
两行之间添加一行新的引用,如下,
#include "DataSequenceHeader.pb.h" #include <google/protobuf/io/coded_stream.h> #include <google/protobuf/io/zero_copy_stream_impl.h>1
2
3
1
2
3
错误二,创建(build)错误
error:
/home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:23:9: error: ‘fast_stage_t’ in ‘class doppia::SoftCascadeOverIntegralChannelsModel’ does not name a type typedef SoftCascadeOverIntegralChannelsModel::fast_stage_t cascade_stage_t; ^ /home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:33:36: error: ‘cascade_stage_t’ was not declared in this scope typedef Cuda::DeviceMemoryLinear2D<cascade_stage_t> gpu_detection_cascade_per_scale_t; ^ /home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:33:51: error: template argument 1 is invalid typedef Cuda::DeviceMemoryLinear2D<cascade_stage_t> gpu_detection_cascade_per_scale_t; ^ /home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:33:86: error: invalid type in declaration before ‘;’ token typedef Cuda::DeviceMemoryLinear2D<cascade_stage_t> gpu_detection_cascade_per_scale_t;1
2
3
4
5
6
7
8
9
10
11
1
2
3
4
5
6
7
8
9
10
11
这里出现错误原因是因为common_settings.cmake中没有添加cuda链接库路径。
solution:
编辑common_settings.cmake,在其中添加一个条件项
elseif(${HOSTNAME} STREQUAL "mx-pc") message(STATUS "Using mx-pc optimisation options") option(USE_GPU "Should the GPU be used ?" TRUE) set(CUDA_BUILD_CUBIN OFF) set(local_CUDA_LIB_DIR "/usr/local/cuda/lib64") set(cuda_LIBS "")1
2
3
4
5
6
7
1
2
3
4
5
6
7
这里mx-pc是我电脑的主机名,你需要将它改成自己电脑的主机名。
错误三,创建(build)错误
error:
/home/mx/doppia/src/objects_detection/SoftCascadeOverIntegralChannelsFastFractionalStage.cpp:24:9: error: ‘swap’ is not a member of ‘std’ std::swap(weak_classifier.level2_true_node, weak_classifier.level2_false_node);1
2
1
2
solution:
在
doppia/src/objects_detection/SoftCascadeOverIntegralChannelsFastFractionalStage.cpp1
1
文件开头添加一行引用
#include<iostream>1
1
解决了以上三个错误后,doppia就可以创建(build)成功啦!但要想运行成功,还得改正以下2个错误。
错误四,链接(link)错误
error:
Linking CXX executable objects_detection /usr/bin/ld: cannot find -lboost_program_options-mt /usr/bin/ld: cannot find -lboost_filesystem-mt /usr/bin/ld: cannot find -lboost_system-mt /usr/bin/ld: cannot find -lboost_thread-mt collect2: error: ld returned 1 exit status make[2]: *** [objects_detection] 错误 1 make[1]: *** [CMakeFiles/objects_detection.dir/all] 错误 2 make: *** [all] 错误 21
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
这里出现错误的原因,是boost库链接出错,这时候我们需要修改CMakeList.txt文件,这里我就直接把CMakeList.txt贴出来,修改的地方做过注释。
solution:
# This is a CMake build file, for more information consult: # http://en.wikipedia.org/wiki/CMake # and # http://www.cmake.org/Wiki/CMake # http://www.cmake.org/cmake/help/syntax.html # http://www.cmake.org/Wiki/CMake_Useful_Variables # http://www.cmake.org/cmake/help/cmake-2-8-docs.html # to compile the local code you can use: cmake ./ && make -j2 # ---------------------------------------------------------------------- # Base CMake setup cmake_minimum_required (VERSION 2.6) set(doppia_root "../../..") set(CMAKE_MODULE_PATH $ENV{CMAKE_MODULE_PATH}) set(CMAKE_MODULE_PATH "./" ${doppia_root} ${CMAKE_MODULE_PATH}) set(CMAKE_MODULE_PATH "/home/rodrigob/work/code/doppia_references/cuda/FindCUDA/CMake/cuda" ${CMAKE_MODULE_PATH}) set(CMAKE_MODULE_PATH "/users/visics/rbenenso/code/references/cuda/FindCUDA/CMake/cuda" ${CMAKE_MODULE_PATH}) # ---------------------------------------------------------------------- # Setup the project include(FindPkgConfig) project (ObjectsDetection) # ---------------------------------------------------------------------- # Site specific configurations include(${doppia_root}/common_settings.cmake) # ---------------------------------------------------------------------- # Setup required libraries pkg_check_modules(libpng REQUIRED libpng) #pkg_check_modules(OpenEXR REQUIRED OpenEXR) pkg_check_modules(opencv REQUIRED opencv>=2.3) #set(vw_LIBRARIES "-lvwCore -lvwImage -lvwStereo -lvwFileIO -lvwMath -lvwInterestPoint") set(opencv_LIBRARIES opencv_core opencv_imgproc opencv_highgui opencv_ml opencv_video opencv_features2d opencv_calib3d #opencv_objdetect opencv_contrib opencv_legacy opencv_flann ) # quick hack for opencv2.4 support # 修改1:find where is boost #(+)find_package(Boost REQUIRED #(+) COMPONENTS program_options filesystem system thread #(+) ) find_package(Boost REQUIRED COMPONENTS program_options filesystem system thread ) # ---------------------------------------------------------------------- # Setup CUDA if(USE_GPU) find_package(CUDA 4.0 REQUIRED) include_directories(${CUDA_INCLUDE_DIRS} ${CUDA_CUT_INCLUDE_DIR}) endif(USE_GPU) # ---------------------------------------------------------------------- # Setup link and include directories set(local_LIBRARY_DIRS "/usr/local/lib" "/users/visics/rbenenso/no_backup/usr/local/lib" "/usr/lib64" "/usr/lib64/atlas" "/usr/lib/sse2/atlas" "/usr/lib/llvm-2.8/lib" ${local_CUDA_LIB_DIR} ) set(local_INCLUDE_DIRS "/users/visics/rbenenso/no_backup/usr/local/include" "/usr/include/eigen2/" "/usr/local/include/eigen2" "/usr/local/cuda/include" ${CUDA_INCLUDE_DIRS} ) link_directories( ${libpng_LIBRARY_DIRS} ${OpenEXR_LIBRARY_DIRS} ${opencv_LIBRARY_DIRS} ${local_LIBRARY_DIRS} ) include_directories( "${doppia_root}/libs" "${doppia_root}/src" ${libpng_INCLUDE_DIRS} ${OpenEXR_INCLUDE_DIRS} ${opencv_INCLUDE_DIRS} ${local_INCLUDE_DIRS} "${doppia_root}/libs/cudatemplates/include" ) if(USE_GPU) cuda_include_directories("${doppia_root}/libs/") endif(USE_GPU) # ---------------------------------------------------------------------- # Collect source files set(doppia_src "${doppia_root}/src") set(doppia_stereo "${doppia_root}/src/stereo_matching") file(GLOB SrcCpp "./ObjectsDetection*.cpp" "./draw*.cpp" "${doppia_src}/*.cpp" #"${doppia_src}/objects_detection/*.c*" "${doppia_src}/objects_detection/Abstract*.c*" "${doppia_src}/objects_detection/*Converter.c*" "${doppia_src}/objects_detection/Base*.c*" "${doppia_src}/objects_detection/*Factory.c*" "${doppia_src}/objects_detection/Greedy*.c*" "${doppia_src}/objects_detection/Detection*.c*" "${doppia_src}/objects_detection/*Model.c*" "${doppia_src}/objects_detection/*Stage.c*" "${doppia_src}/objects_detection/*Integral*.c*" "${doppia_src}/objects_detection/MultiscalesIntegral*.c*" "${doppia_src}/objects_detection/integral_channels/Integral*.cpp" "${doppia_src}/objects_detection/FastestPedestrian*.c*" "${doppia_src}/objects_detection/DetectorSearchRange.c*" "${doppia_src}/objects_detection/*.pb.c*" "${doppia_src}/objects_detection/non_maximal_suppression/*.c*" "${doppia_src}/objects_tracking/*.cpp" "${doppia_src}/applications/*.cpp" "${doppia_src}/applications/stixel_world/*Gui.cpp" "${doppia_src}/applications/stixel_world/draw*.cpp" #"${doppia_stereo}/*.cpp" "${doppia_stereo}/cost_volume/*CostVolume.cpp" "${doppia_stereo}/cost_volume/*CostVolumeEstimator*.cpp" "${doppia_stereo}/cost_volume/DisparityCostVolumeFromDepthMap.cpp" "${doppia_stereo}/cost_functions.cpp" "${doppia_stereo}/CensusCostFunction.cpp" "${doppia_stereo}/CensusTransform.cpp" "${doppia_stereo}/GradientTransform.cpp" "${doppia_stereo}/AbstractStereoMatcher.cpp" "${doppia_stereo}/AbstractStereoBlockMatcher.cpp" "${doppia_stereo}/SimpleBlockMatcher.cpp" "${doppia_stereo}/MutualInformationCostFunction.cpp" "${doppia_stereo}/ConstantSpaceBeliefPropagation.cpp" "${doppia_stereo}/qingxiong_yang/*.cpp" "${doppia_stereo}/SimpleTreesOptimizationStereo.cpp" "${doppia_stereo}/OpenCvStereo.cpp" "${doppia_stereo}/ground_plane/*.cpp" "${doppia_stereo}/stixels/*.cpp" #"${doppia_stereo}/stixels/*.cc" "${doppia_src}/video_input/*.cpp" "${doppia_src}/video_input/calibration/*.c*" "${doppia_src}/video_input/preprocessing/*.cpp" #"${doppia_src}/features_tracking/*.cpp" "${doppia_src}/image_processing/*.cpp" "${doppia_src}/drawing/gil/*.cpp" ) file(GLOB HelpersCpp #"${doppia_src}/helpers/*.cpp" "${doppia_src}/helpers/data/*.c*" "${doppia_src}/helpers/any_to_string.cpp" "${doppia_src}/helpers/get_section_options.cpp" "${doppia_src}/helpers/Log.cpp" "${doppia_src}/helpers/loggers.cpp" "${doppia_src}/helpers/AlignedImage.cpp" "${doppia_src}/helpers/replace_environment_variables.cpp" "${doppia_src}/helpers/objects_detection/*.cpp" ) file(GLOB SrcGpuCpp "${doppia_src}/objects_detection/Gpu*.cpp" "${doppia_src}/objects_detection/integral_channels/Gpu*.cpp" "${doppia_src}/helpers/gpu/*.cpp" #"${doppia_stereo}/SimpleTreesGpuStereo.cpp" ) file(GLOB SrcCuda "${doppia_src}/objects_detection/integral_channels/gpu/*.cu" "${doppia_src}/objects_detection/integral_channels/gpu/*.cpp" "${doppia_src}/objects_detection/gpu/*.cu" "${doppia_src}/objects_detection/gpu/*.cpp" #"${doppia_src}/helpers/gpu/*.cu" # "${doppia_stereo}/*.cu.c*" # "${doppia_stereo}/*.cu" # "${doppia_stereo}/gpu/*.cu.c*" # "${doppia_stereo}/gpu/*.cu" ) list(REMOVE_ITEM SrcCpp ${SrcCuda}) # just in case if(USE_GPU) # add GPU related source code to the executable list list(APPEND SrcCpp ${SrcGpuCpp}) # add GPU related libraries list(APPEND opencv_LIBRARIES opencv_gpu) # ---------------------------------------------------------------------- # Compile CUDA stuff cuda_include_directories(${local_CUDA_CUT_INCLUDE_DIRS}) cuda_include_directories(${CUDA_INCLUDE_DIRS} ${CUDA_CUT_INCLUDE_DIR} ${local_CUDA_CUT_INCLUDE_DIR}) link_directories(${local_CUDA_CUT_LIBRARY_DIRS}) cuda_add_library(cuda_stuff_library ${SrcCuda}) target_link_libraries(cuda_stuff_library ${CUDA_LIBRARIES} ${cutil_LIB} ) #set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} --generate-line-info) # used during profiling endif(USE_GPU) # ---------------------------------------------------------------------- # Create the executable #set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++0x") # required for unrestricted unions #set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -p") # add gprof information add_library(cpp_stuff_library ${SrcCpp} ${HelpersCpp}) add_executable(objects_detection "./objects_detection.cpp") target_link_libraries(objects_detection cpp_stuff_library ${cg_LIBRARIES} # linking with CgGL _after_ boost_program_options generates a segmentation fault ! boost_program_options 1.39 has a bug #修改2:link to boost #(-)boost_program_options-mt boost_filesystem-mt boost_system-mt boost_thread-mt #(+)${Boost_LIBRARIES} ${Boost_LIBRARIES} protobuf pthread SDL X11 Xext #Xrandr gomp ${libpng_LIBRARIES} jpeg # ${OpenEXR_LIBRARIES} ${opencv_LIBRARIES} #${vw_LIBRARIES} #csparse sparse spblas mv #lapack blas atlas ${google_perftools_LIBS} # enables profiling, see http://code.google.com/p/google-perftools #`OcelotConfig -l` #ocelot #boost_system-mt boost_filesystem-mt boost_thread-mt #GLEW #LLVMAsmParser LLVMX86Disassembler LLVMX86AsmParser LLVMX86CodeGen LLVMSelectionDAG #LLVMAsmPrinter LLVMMCParser LLVMX86AsmPrinter LLVMX86Info LLVMJIT #LLVMExecutionEngine LLVMCodeGen LLVMScalarOpts LLVMInstCombine LLVMTransformUtils LLVMipa #LLVMAnalysis LLVMTarget LLVMMC LLVMCore LLVMSupport LLVMSystem ) if(USE_GPU) target_link_libraries(objects_detection cuda_stuff_library ${local_CUDA_LIB}) endif(USE_GPU) # ----------------------------------------------------------------------1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
到这里就OK啦!
最后运行成功你会看到一个简短地video,以及下面这样的信息
2015-05-09 16:33:23 {7feb06974880} [ BaseIntegralChannelsDetector ] : Warning: At scale index 47 the detection window size is larger than the biggest ground plane corridor. Setting the detection search to a single line. scale_index == 47, original_height == 18, updated_height == 1 Expected speed gain == 5.28x (num pixels original/updated) GpuVeryFastIntegralChannelsDetector::compute_v2 max search range (min_x, min_y; max_x, max_y) == (0, 0; 153, 58) 2015-05-09 16:33:23 {7feb06974880} [ GpuIntegralChannelsDetector ] : scaled_x == 640, scaled_y == 480 Requested frame number 11 but frames should be in range (0, 10) Processed a total of 10 input frames Average objects detection speed per iteration 29.36 [Hz] (in the last 10 iterations) End of game, have a nice day.1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
好啦,doppia编译到此也就结束啦!
希望这几篇文章能帮助正在读博客的你。
doppia及作者相关介绍链接:
http://blog.csdn.net/xizero00/article/details/43227019
https://bitbucket.org/rodrigob/doppia
相关文章推荐
- 【视频开发】【计算机视觉】doppia编译之一:前言及安装CUDA
- 【视频开发】【计算机视觉】doppia编译之二:boost安装
- 【视频开发】【计算机视觉】doppia编译之三:编译安装opencv库
- doppia编译之四:安装其他库、编译和运行doppia
- 一个关于VS2008在其他未安装VS2008的电脑上无法运行编译的程序的问题 收藏
- eclipse中配置hadoop开发环境-----删除之前版本的hadoop插件、编译hadoop eclipse插件、安装插件、eclipse下运行hadoop程序
- 【视频开发】【计算机视觉】相机标定(Camera calibration)原理、步骤
- MyEclipse 6 实战开发讲解视频入门 0: 下载 安装 运行 HelloWorld
- vs 项目编译运行报错:无法启动此程序,因为计算机中丢失 osg130-osgDB.dll。尝试重新安装该程序以解决此问题
- 【计算机视觉】【视频开发】遗留物检测中的物体遗留和移除检测
- 一个关于VS2008在其他未安装VS2008的电脑上无法运行编译的程序的问题
- 【Qt开发】【计算机视觉】OpenCV在Qt-MinGw下的编译库
- 2.建立一个Win32 Console Application工程,编译好的exe文件在其他的电脑直接运行而不需要安装任何其他的package.
- MyEclipse 6 实战开发讲解视频入门 0: 下载 安装 运行 HelloWorld
- MyEclipse 6 实战开发讲解视频入门 0: 下载 安装 运行 HelloWorld
- 【VS开发】【计算机视觉】使用opencv静态库编译程序时爆出错误的解决方案
- 【视频开发】【计算机视觉】全景视频拼接关键技术
- 【计算机视觉】【ARM-Linux开发】Ubuntu14.04下安装opencv3(anaconda3 Python3.6)
- [计算机视觉][ARM-Linux开发] Ubuntu14.04安装OpenCV3.2中遇到的问题的解决方案
- 【计算机视觉】【视频开发】智能视频监控中的遗留物或搬移物检测