lppllppl920/sage-slam

Official repo for the paper "SAGE: SLAM with Appearance and Geometry Prior for Endoscopy" (ICRA 2022)

sage-slam's Issues

Training on Custom Video

Hello,

I am attempting to run training on my own endoscopy video sequence. I see that the given dataset already has groundtruth as well as hdf5 file for the data. In addition, the training code already assumes this data format. What would you recommend to do to run training and the system from my own image sequence? Thank you.

BOW model

Hi auther~I wonder if I need to train the Bag-Of-Word model on my datasets ?

ValueError: a must be non-empty?

There was an error running step 6.
why?
Looking forward to the author's reply!!!

Step 9: There is no build file after running

Hello dear author
When I ran to step 9 according to the steps you provided (of which steps 6, 7, and 8 I skipped directly without running), the error "Building TEASER-plusplus"
make: *** No rule to make target 'install'. Stop.",
and after using tree -d -L 3, the build folder did not appear in the project file directory , and where the command for step nine I was running in the docker desktop terminal

just run monolithic depth estimation

Hello author, I now have my own monocular dataset, which is also the scene of the channel, I just want to do monolithic depth estimation (I just want to generate a depth map like yours), which program should I run or refer to, or what are my execution steps, thank you author

Generate fly through surface reconstructions

I do see in the github repo that fly-through surface reconstruction results are present, and I am trying to generate the same. I understand that the script - 'generate_reconstruction_fly_through.py' should be used, but it requires a lot of input files that I am not entirely sure how to generate. It would be very useful if you can show an example of how to use the script.

about the data

hello, the data bag_1 is the full dataset? if not, could you give me the whole dataset? there are only about 1000 pictures, I want to get all to train other model thank you?

TypeError: can't pickle cv2.FastFeatureDetector objects

I encountered this error in the 280 line of trainning.py

for curr_iter, input_dict in enumerate(train_loader):

I'm not sure if it is caused by the OpenCV version.

Sophus ensure failed in function

hello dear author,
I followed your steps but when I tried to test run the SAGE-SLAM system on the example dataset, I got this

I0410 12:37:23.335498   486 main.cpp:370] Logging directory: /tmp/SAGE-SLAM__start_002603_end_002984_stride_1000_segment_00_2022-04-10-12:37:23
I0410 12:37:26.994796   490 deepfactors.cpp:1251] [DeepFactors<Scalar, CS>::MappingBackend] Mapping update thread started
I0410 12:37:26.995555   486 live_demo.cpp:147] [LiveDemo<CS>::ProcessingLoop] Entering processing loop
I0410 12:37:26.995613   486 live_demo.cpp:161] [LiveDemo<CS>::ProcessingLoop] Initializing system on the first frame
I0410 12:37:26.997876   486 live_demo.cpp:164] [LiveDemo<CS>::ProcessingLoop] Bootstrap frame 0(0)
I0410 12:37:34.324018   486 live_demo.cpp:206] [LiveDemo<CS>::ProcessingLoop] Process frame 1(1)
Sophus ensure failed in function 'Sophus::SO3<Scalar_, Options>::SO3(const Transformation&) [with Scalar_ = float; int Options = 0; Sophus::SO3<Scalar_, Options>::Transformation = Eigen::Matrix<float, 3, 3>]', file '/home/pwn20tty/system/thirdparty/install_Debug/include/sophus/so3.hpp', line 472.
R is not orthogonal:
      1.00001  7.31088e-08  1.09612e-07
 7.31088e-08      1.00001 -4.67872e-08
 1.09612e-07 -4.67872e-08            1
signal 6 (Aborted), address is 0x3e8000001e6 from 0x7fb75a6bf00b
[bt]: (1) /usr/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcb) [0x7fb75a6bf00b]
[bt]: (2) /usr/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcb) [0x7fb75a6bf00b]
[bt]: (3) /usr/lib/x86_64-linux-gnu/libc.so.6(abort+0x12b) [0x7fb75a69e859]
[bt]: (4) /home/pwn20tty/build/Debug/bin/df_demo(_ZN6Sophus13defaultEnsureIJKN5Eigen7ProductINS1_6MatrixIfLi3ELi3ELi0ELi3ELi3EEENS1_9TransposeIKS4_EELi0EEEEEEvPKcSB_iSB_DpOT_+0xa9) [0x5648b1be56e4]
[bt]: (5) /home/pwn20tty/build/Debug/bin/df_demo(_ZN6Sophus3SO3IfLi0EEC1ERKN5Eigen6MatrixIfLi3ELi3ELi0ELi3ELi3EEE+0x9b) [0x5648b1bdbbff]
[bt]: (6) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df13CameraTracker13TrackNewFrameERNS_5FrameIfEEbbbbb+0x2ecc) [0x7fb7ac94632c]
[bt]: (7) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df11DeepFactorsIfLi16EE10TrackFrameEv+0x5c1) [0x7fb7acaacbbb]
[bt]: (8) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df11DeepFactorsIfLi16EE12ProcessFrameEdRKN2cv3MatE+0x26e) [0x7fb7acaa7f0c]
[bt]: (9) /home/pwn20tty/build/Debug/bin/df_demo(+0x1931af) [0x5648b1bd21af]
[bt]: (10) /home/pwn20tty/build/Debug/bin/df_demo(+0x1918b1) [0x5648b1bd08b1]
[bt]: (11) /home/pwn20tty/build/Debug/bin/df_demo(main+0xfe3) [0x5648b1c449cf]
[bt]: (12) /usr/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf3) [0x7fb75a6a0083]
[bt]: (13) /home/pwn20tty/build/Debug/bin/df_demo(+0x18995e) [0x5648b1bc895e]

I am a noob. I have no idea about how to solve this problem.
Could you give me some instructions？

How to solve the dependency of generate_reconstruction_fly_through.py?

I have run and get some result of the SLAM, and I want to generate surface. But I found when I run the generate_reconstruction_fly_through.py, there always have dependency error.
And I found that I need to install a in-house version of the meshrender, and I only find a link in your another repo DenseReconstruction-Pytorch, but I don't know whether it is right. But I found there still an error need to be solved.
The first is

ImportError: cannot import name 'RenderMode' from 'perception'

I try to solve this by convert perception to autolab_core or meshpy, but it is no use. Can you help me by show me your dependency about these package?
Thanks!

hdf5 data

hello dear author,
I'd like to know how to obtain the hdf5 data, cause I didn't find it in your google drive. could you give me some instructions?

std_depth && render_depth

HelloSorry to bother youI want to know the difference between std_depth and render_depth in the given datasets, besides, how can I acquire them?
Thank you~

OpenGL error: GL_INVALID_ENUM in glCreateShader(GL_GEOMETRY_SHADER)

Thanks for trying to solve my problem! (Sophus ensure failed in function)
I exported the docker container and copied it to another computer.And now it can run successfully. (That's so weird.)

Then I got a new problem……
I set '--enable_gui=true' and I got

I0418 13:35:04.774458   200 main.cpp:370] Logging directory: /tmp/SAGE-SLAM__start_002603_end_002984_stride_1000_segment_00_2022-04-18-13:35:04
I0418 13:35:06.797400   205 deepfactors.cpp:1251] [DeepFactors<Scalar, CS>::MappingBackend] Mapping update thread started
I0418 13:35:06.797688   200 live_demo.cpp:147] [LiveDemo<CS>::ProcessingLoop] Entering processing loop
I0418 13:35:06.797719   200 live_demo.cpp:161] [LiveDemo<CS>::ProcessingLoop] Initializing system on the first frame
I0418 13:35:06.798633   200 live_demo.cpp:164] [LiveDemo<CS>::ProcessingLoop] Bootstrap frame 0(0)
Framebuffer with requested attributes not available. Using available framebuffer. You may see visual artifacts.I0418 13:35:06.912933   206 visualizer.cpp:46] [OglDebugCallback] OpenGL Debug message: GL_INVALID_ENUM in glCreateShader(GL_GEOMETRY_SHADER)
F0418 13:35:06.912987   206 visualizer.cpp:49] [OglDebugCallback] OpenGL error: GL_INVALID_ENUM in glCreateShader(GL_GEOMETRY_SHADER)
*** Check failure stack trace: ***
    @     0x7f43f5ef21c3  google::LogMessage::Fail()
    @     0x7f43f5ef725b  google::LogMessage::SendToLog()
    @     0x7f43f5ef1ebf  google::LogMessage::Flush()
    @     0x7f43f5ef26ef  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f43f639177c  df::OglDebugCallback()
    @     0x7f431bb17ffa  (unknown)
    @     0x7f431bba9c54  (unknown)
    @     0x7f43f643d68b  pangolin::GlSlProgram::AddPreprocessedShader()
    @     0x7f43f643e6a6  pangolin::GlSlProgram::AddShaderFile()
    @     0x7f43f643e965  pangolin::GlSlProgram::AddShaderFromFile()
    @     0x7f43f643bcbb  df::KeyframeRenderer::Init()
    @     0x7f43f6391c58  df::Visualizer::Init()
    @     0x556b4e10189a  df::LiveDemo<>::VisualizerLoop()
    @     0x556b4e1721d0  std::__invoke_impl<>()
    @     0x556b4e17212f  std::__invoke<>()
    @     0x556b4e172099  _ZNSt5_BindIFMN2df8LiveDemoILi16EEEFvvEPS2_EE6__callIvJEJLm0EEEET_OSt5tupleIJDpT0_EESt12_Index_tupleIJXspT1_EEE
    @     0x556b4e17201d  std::_Bind<>::operator()<>()
    @     0x556b4e171fc4  std::__invoke_impl<>()
    @     0x556b4e171f6d  std::__invoke<>()
    @     0x556b4e171f0e  _ZNSt6thread8_InvokerISt5tupleIJSt5_BindIFMN2df8LiveDemoILi16EEEFvvEPS5_EEEEE9_M_invokeIJLm0EEEEvSt12_Index_tupleIJXspT_EEE
    @     0x556b4e171ee0  std::thread::_Invoker<>::operator()()
    @     0x556b4e1714e0  std::thread::_State_impl<>::_M_run()
    @     0x7f43a2ea219d  execute_native_thread_routine
    @     0x7f43f5957609  start_thread
    @     0x7f43a2cf1133  clone
signal 6 (Aborted), address is 0x3e8000000c8 from 0x7f43a2c1500b
[bt]: (1) /usr/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcb) [0x7f43a2c1500b]
[bt]: (2) /usr/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcb) [0x7f43a2c1500b]
[bt]: (3) /usr/lib/x86_64-linux-gnu/libc.so.6(abort+0x12b) [0x7f43a2bf4859]
[bt]: (4) /usr/lib/x86_64-linux-gnu/libglog.so.0(+0xa90e) [0x7f43f5eef90e]
[bt]: (5) /usr/lib/x86_64-linux-gnu/libglog.so.0(+0xd1c3) [0x7f43f5ef21c3]
[bt]: (6) /usr/lib/x86_64-linux-gnu/libglog.so.0(_ZN6google10LogMessage9SendToLogEv+0x26b) [0x7f43f5ef725b]
[bt]: (7) /usr/lib/x86_64-linux-gnu/libglog.so.0(_ZN6google10LogMessage5FlushEv+0xbf) [0x7f43f5ef1ebf]
[bt]: (8) /usr/lib/x86_64-linux-gnu/libglog.so.0(_ZN6google15LogMessageFatalD2Ev+0xf) [0x7f43f5ef26ef]
[bt]: (9) /home/pwn20tty/build/Debug/sources/gui/libdf_gui.so(_ZN2df16OglDebugCallbackEjjjjiPKcPKv+0x182) [0x7f43f639177c]
[bt]: (10) /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so(+0x310ffa) [0x7f431bb17ffa]
[bt]: (11) /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so(+0x3a2c54) [0x7f431bba9c54]
[bt]: (12) /home/pwn20tty/build/Debug/sources/gui/libdf_gui.so(_ZN8pangolin11GlSlProgram21AddPreprocessedShaderENS_14GlSlShaderTypeERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES9_+0x59) [0x7f43f643d68b]
[bt]: (13) /home/pwn20tty/build/Debug/sources/gui/libdf_gui.so(_ZN8pangolin11GlSlProgram13AddShaderFileERKNS0_16ShaderFileOrCodeE+0x460) [0x7f43f643e6a6]
[bt]: (14) /home/pwn20tty/build/Debug/sources/gui/libdf_gui.so(_ZN8pangolin11GlSlProgram17AddShaderFromFileENS_14GlSlShaderTypeERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt3mapIS7_S7_St4lessIS7_ESaISt4pairIS8_S7_EEERKSt6vectorIS7_SaIS7_EE+0xe9) [0x7f43f643e965]
[bt]: (15) /home/pwn20tty/build/Debug/sources/gui/libdf_gui.so(_ZN2df16KeyframeRenderer4InitERKNS_13PinholeCameraIfEE+0x2d1) [0x7f43f643bcbb]
[bt]: (16) /home/pwn20tty/build/Debug/sources/gui/libdf_gui.so(_ZN2df10Visualizer4InitERKNS_13PinholeCameraIfEENS_18DeepFactorsOptionsE+0x150) [0x7f43f6391c58]
[bt]: (17) /home/pwn20tty/build/Debug/bin/df_demo(+0x19389a) [0x556b4e10189a]
[bt]: (18) /home/pwn20tty/build/Debug/bin/df_demo(+0x2041d0) [0x556b4e1721d0]
[bt]: (19) /home/pwn20tty/build/Debug/bin/df_demo(+0x20412f) [0x556b4e17212f]
[bt]: (20) /home/pwn20tty/build/Debug/bin/df_demo(+0x204099) [0x556b4e172099]
[bt]: (21) /home/pwn20tty/build/Debug/bin/df_demo(+0x20401d) [0x556b4e17201d]
[bt]: (22) /home/pwn20tty/build/Debug/bin/df_demo(+0x203fc4) [0x556b4e171fc4]
[bt]: (23) /home/pwn20tty/build/Debug/bin/df_demo(+0x203f6d) [0x556b4e171f6d]
[bt]: (24) /home/pwn20tty/build/Debug/bin/df_demo(+0x203f0e) [0x556b4e171f0e]
[bt]: (25) /home/pwn20tty/build/Debug/bin/df_demo(+0x203ee0) [0x556b4e171ee0]
[bt]: (26) /home/pwn20tty/build/Debug/bin/df_demo(+0x2034e0) [0x556b4e1714e0]
[bt]: (27) /opt/conda/lib/libstdc++.so.6(+0xc819d) [0x7f43a2ea219d]
[bt]: (28) /usr/lib/x86_64-linux-gnu/libpthread.so.0(+0x8609) [0x7f43f5957609]
[bt]: (29) /usr/lib/x86_64-linux-gnu/libc.so.6(clone+0x43) [0x7f43a2cf1133]
signal 11 (Segmentation fault), address is (nil) from 0x7f43ede8753e
[bt]: (1) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZNKSt10_HashtableINSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEESt4pairIKS5_mESaIS8_ENSt8__detail10_Select1stESt8equal_toIS5_ESt4hashIS5_ENSA_18_Mod_range_hashingENSA_20_Default_ranged_hashENSA_20_Prime_rehash_policyENSA_17_Hashtable_traitsILb1ELb0ELb1EEEE19_M_find_before_nodeEmRS7_m+0x2e) [0x7f43ede8753e]
[bt]: (2) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZNKSt10_HashtableINSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEESt4pairIKS5_mESaIS8_ENSt8__detail10_Select1stESt8equal_toIS5_ESt4hashIS5_ENSA_18_Mod_range_hashingENSA_20_Default_ranged_hashENSA_20_Prime_rehash_policyENSA_17_Hashtable_traitsILb1ELb0ELb1EEEE19_M_find_before_nodeEmRS7_m+0x2e) [0x7f43ede8753e]
[bt]: (3) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit10is_enabledEPKcNS0_16JitLoggingLevelsE+0x1c8) [0x7f43ede9f4c8]
[bt]: (4) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit6InlineERNS0_5GraphE+0x39) [0x7f43edf248f9]
[bt]: (5) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit16preoptimizeGraphERSt10shared_ptrINS0_5GraphEE+0x10) [0x7f43edd62ad0]
[bt]: (6) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(+0x3869559) [0x7f43edd64559]
[bt]: (7) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(+0x386b18b) [0x7f43edd6618b]
[bt]: (8) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit13GraphFunction3runERSt6vectorIN3c106IValueESaIS4_EE+0xe) [0x7f43edd628ee]
[bt]: (9) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(+0x3b0effd) [0x7f43ee009ffd]
[bt]: (10) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit16InterpreterState3runERSt6vectorIN3c106IValueESaIS4_EE+0x30) [0x7f43edff7630]
[bt]: (11) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(+0x3aef534) [0x7f43edfea534]
[bt]: (12) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit13GraphFunctionclESt6vectorIN3c106IValueESaIS4_EERKSt13unordered_mapINSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES4_St4hashISD_ESt8equal_toISD_ESaISt4pairIKSD_S4_EEE+0x3e) [0x7f43edd62d9e]
[bt]: (13) /opt/conda/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so(_ZN5torch3jit6MethodclESt6vectorIN3c106IValueESaIS4_EERKSt13unordered_mapINSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES4_St4hashISD_ESt8equal_toISD_ESaISt4pairIKSD_S4_EEE+0x168) [0x7f43edd72748]
[bt]: (14) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN5torch3jit6Module7forwardESt6vectorIN3c106IValueESaIS4_EE+0x103) [0x7f43f4ff5065]
[bt]: (15) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df14FeatureNetwork19GenerateFeatureMapsEN2at6TensorES2_RS2_S3_+0x2e1) [0x7f43f4ff843f]
[bt]: (16) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df6MapperIfLi16EE13BuildKeyframeEdRKN2cv3MatERKN6Sophus3SE3IfLi0EEE+0xcde) [0x7f43f4fa1066]
[bt]: (17) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df6MapperIfLi16EE12InitOneFrameEdRKN2cv3MatE+0x96) [0x7f43f4f977bc]
[bt]: (18) /home/pwn20tty/build/Debug/sources/core/libdf_core.so(_ZN2df11DeepFactorsIfLi16EE17BootstrapOneFrameEdRKN2cv3MatE+0x101) [0x7f43f5000019]
[bt]: (19) /home/pwn20tty/build/Debug/bin/df_demo(+0x192b26) [0x556b4e100b26]
[bt]: (20) /home/pwn20tty/build/Debug/bin/df_demo(+0x1918b1) [0x556b4e0ff8b1]
[bt]: (21) /home/pwn20tty/build/Debug/bin/df_demo(main+0xfe3) [0x556b4e1739cf]
[bt]: (22) /usr/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf3) [0x7f43a2bf6083]
[bt]: (23) /home/pwn20tty/build/Debug/bin/df_demo(+0x18995e) [0x556b4e0f795e]

I am sure that X11 display is working normally. I can run firefox within the docker.
Did this only happening on my machine?
Could you please help me to solve this?