Toytiny/CMFlow

Issues

• VoD preprocessing results in gt_mask being only 1.0

  #16 opened 3 months ago by LeonSchw
  5

• I observed that gt_labels mostly contain zeros, and when I calculate pos1 + label, it does not equal pos2 as expected. I suspect that there might be an issue with the ground truth data or how I’m handling the transformation.

  #15 opened 4 months ago by Wangyiheng962
  2

• About visualization

  #5 opened 2 years ago by xlbjy
  2

• The center param in get_bbx_param() function

  #14 opened 6 months ago by MMOCKING
  1

• When running "python preprocess/preprocess_vod.py",I did not create"label_track_gt",but it can run

  #13 opened 10 months ago by KNzzzzzzzzzz
  6

• About discontinuous frames

  #12 opened a year ago by EricLiuhhh
  2

• Vehicle Ego Velocity

  #11 opened a year ago by paaraujo
  1

• cannot import name '_nt_quote_args' from 'distutils.spawn' (D:\Anaconda\envs\Hi\lib\site-packages\setuptools\_distutils\spawn.py)

  #10 opened a year ago by Random-junhao
  2

• false request

  #9 opened a year ago by Random-junhao
  0

• Difference on performance

  #8 opened a year ago by FengyeXZ
  4

• About the trained CMFlow (T) model trained with extra unannotated data

  #7 opened a year ago by xlbjy
  1

• testing

  #6 opened a year ago by JiaNing-Z
  4

• radar frame

  #4 opened 2 years ago by lushouyi
  1

• Visualisation tool to visualise the .json file in the results from test. I would like to see some radar points cast onto the image.

  #2 opened 2 years ago by Toytiny
  1

• But when I rerun the code, I get a RuntimeError:CUDA out of memory. I changed the size of the batch size in the RAFT/ train.py file, but it doesn't fix the problem

  #3 opened 2 years ago by Toytiny
  2

• When running "python preprocess/preprocess_vod.py", An error has occurred"No module named 'utils.common'"

  #1 opened 2 years ago by syh1620
  15