yuxng / DA-RNN

Semantic Mapping with Data Associated Recurrent Neural Networks
MIT License
170 stars 72 forks source link

Error when make kinect_fusion (execute sh make.sh successfully) #8

Closed JackHenry1992 closed 7 years ago

JackHenry1992 commented 7 years ago

Thks for sharing your great job. I've encoutered error when built kinect_fusion prj.

image


It seems that sophus not found , but in my /usr/local/include/sophus exist headers of sophus and I've added this path to cmakelist.txt image

I have no idea how to deal with this error, could you give me some tips?

kevinkit commented 7 years ago

see #2 looks like you have the wrong sophus version, since the SO3 group was changed. Make sure that your sh-id matches after checking out the repo. I have encountered the same problem and could fix it like this

kevinkit commented 7 years ago

did you get it to run or have you met other errors?

Am 13.07.2017 14:22 schrieb "JackHenry1992" notifications@github.com:

Closed #8 https://github.com/yuxng/DA-RNN/issues/8.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/yuxng/DA-RNN/issues/8#event-1162459512, or mute the thread https://github.com/notifications/unsubscribe-auth/AGhecVKBOBzmqL2b4wHiAK_B6rkMzMHIks5sNgv6gaJpZM4OW5YR .

JackHenry1992 commented 7 years ago

If built libkfusion.so successfully, image but encountered pangolin error when building kinect_fusion.cpp (same with #7 ), will this influence DA-RNN system run if I don't need display result ? I run DA-RNN in docker as @kevinkit.

kevinkit commented 7 years ago

your screenshot looks like windows, are you on an ubuntu machine?

training etc. should work without kinect_fusion, however it may be hard to test the data. The scripts are provided in the repo

if you have an other way to test the output it'd be great if you may want to share the code :)

Am 13.07.2017 15:25 schrieb "JackHenry1992" notifications@github.com:

If built libkfusion.so successfully, [image: image] https://user-images.githubusercontent.com/9390228/28168335-57ac35aa-6811-11e7-8df4-e59fb37f2ae3.png but encountered pangolin error when building kinect_fusion.cpp (same with

7 https://github.com/yuxng/DA-RNN/issues/7 ), will this influence

DA-RNN system run if I don't need display result ? I run DA-RNN in docker as @kevinkit https://github.com/kevinkit.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/yuxng/DA-RNN/issues/8#issuecomment-315076884, or mute the thread https://github.com/notifications/unsubscribe-auth/AGhecVnUe5zgM_jmbpCoEWs4n3yrYjjZks5sNhrggaJpZM4OW5YR .

JackHenry1992 commented 7 years ago

This is my xftp screenshot. I build DA-RNN in server docker image like #7 , and encountered the same error, according to #7, it seems not easy to build kinect_fusion successfully , I have no idea now.

Now compile the cython interface step ok. So only one error exist (pangolin of kinect_fusion.cpp)

jianxiongcai commented 5 years ago

I met exactly the same problem after changing the Sophus version to the one mentioned in #2. Using the latest Sophus version (sha1: 26c200265e2eb3d76e5ab00a99ada686d6a80d15) works for me :)

Environment: Ubuntu 16.04 with cuda 8.0

Dinghow commented 5 years ago

I met exactly the same problem after changing the Sophus version to the one mentioned in #2. Using the latest Sophus version (sha1: 26c200265e2eb3d76e5ab00a99ada686d6a80d15) works for me :)

Environment: Ubuntu 16.04 with cuda 8.0

Thanks a lot! It really helps! BTW, to install the latest Sophus, I updated my eigen to 3.3.1