search

VisualComputingInstitute / mots_tools

Tools for evaluating and visualizing results for the Multi Object Tracking and Segmentation (MOTS) task
MIT License
222 stars 45 forks source link

Evaluating a tracking result #3

Closed SamihaSara closed 4 years ago

SamihaSara commented 4 years ago

I am trying to understand the evaluation codes. I have not run the track-RCNN yet. I am facing the following error While trying to run. Please suggest me some solutions to this error. Can anyone please tell how to input seqmap ? I have not changed the seqmap files from mots_eval.

Capture

pvoigtlaender commented 4 years ago

Hi,

can you please show the call of the script including the passed command line arguments?

SamihaSara commented 4 years ago

Thanks for replying. I am in the Evaluating a tracking result section. Here is the command and then the error messages I am getting.

Capture

Capture

Here seqmap file is exactly copied from fulltrain.seqmap , gt_folder has Annotations in txt format and tracking_results is copied from TrackR-CNN tracking result. I have tried with Annotations in png format but getting the same error. Please let me know what I am doing wrong

SamihaSara commented 4 years ago

After adding a breakpoint in io.py (from mots_common) here is the output after print()

Capture

pvoigtlaender commented 4 years ago

We resolved this issue via email.

jsh-me commented 4 years ago

Hello I have the same problem. Can you tell me what you did to fix it? I hope to solve this problem:( Thanks.

pvoigtlaender commented 4 years ago

Hi @jsh-me Are you trying to eval MOTSChallenge or KITTI MOTS?