Empty detections and Assertion Error while running Forwarding and Tracking

[devdut1999]

I'm running the code on google colab

1. Run python main.py configs/conv3d_sep2 "{"task":"forward_tracking","dataset":"KITTI_segtrack_feed","load_epoch_no":5,"batch_size":5,"export_detections":true,"do_tracking":false,"video_tags_to_load":["0002","0006","0007","0008","0010","0013","0014","0016","0018","0000","0001","0003","0004","0005","0009","0011","0012","0015","0017","0019","0020"]}"

This command gives empty detections in the forwarding directory that is the generated detection text files in TrackR-CNN/forwarded/conv3d_sep2/detections/5 are of 0 bytes.

2. Run python main.py configs/conv3d_sep2"{"build_networks":false,"import_detections":true,"task":"forward_tracking","dataset":"KITTI_segtrack_feed","do_tracking":true,"visualize_detections":true,"visualize_tracks":false,"load_epoch_no":5,"video_tags_to_load":["0002","0006","0007","0008","0010","0013","0014","0016","0018"]}" gives the following error : File "/content/drive/MyDrive/TrackR-CNN/datasets/Dataset.py", line 252, in _load_inputfile_lists assert len(l) > 0 AssertionError {"build_networks":false,"import_detections":true,"task":"forward_tracking","dataset":"KITTI_segtrack_feed","do_tracking":true,"visualize_detections":false,"visualize_tracks":false,"load_epoch_no":5,"video_tags_to_load":["0002","0006","0007","0008","0010","0013","0014","0016","0018"]}

Running step 2 with the detections given by the TrackRCNN team on their official website also gives the assertion error.

My directory structure and configs/conv3d_sep2 is the exact same as mentioned in the README.md I've also changed the default path in the datasets/KITTI/segtrack/KITTI_segtrack.py .

Still I'm not able to solve the issue. Please help me solve this issue

[pvoigtlaender]

Sorry, from this explanation I can't see what causes the problem. It could very well be some problem with paths, that some data is not found. You can try to use a debugger to narrow down the problem. Or you could try to first get it to run locally if you have a proper GPU.