I can't get the CVIDF1 and CVMA indicators.

[wwppnn]

Hello author, thank you and your team for the outstanding contribution. I encountered an issue while reproducing DIVOTrack, which is that I cannot obtain the CVIDF1 and CVMA metrics through the evaluation code. All the results obtained after running the evaluation code are related metrics for single-object tracking such as MOTA and IDF1. I have searched through the code you provided using keywords but did not find any relevant code for evaluating CVIDF1 and CVMA metrics. Do you know what could be the issue? Thank you very much for your assistance. In the README, you mentioned that these two metrics are obtained in MOTChallengeEvalKit_cv_test, but I couldn’t find the related code. In the code snippet below, CVMA metric is mentioned, but it seems that the two places indicated by the red arrows are saving the same metric, so I am still unable to obtain the aforementioned metrics. Do you know how to resolve this? Thank you .

[Hank0626]

Please refer to this file https://github.com/shengyuhao/DIVOTrack/blob/main/MOTChallengeEvalKit_cv_test/command.txt

[chen-si-jia]

Please refer to this file https://github.com/shengyuhao/DIVOTrack/blob/main/MOTChallengeEvalKit_cv_test/command.txt

Using command "python MOT/evalMOT.py", the result is as follows:

CVIDF1 and CVMA indicators are not included in the result.

How to get CVIDF1 and CVMA indicators?

[Hank0626]

Please refer to this file https://github.com/shengyuhao/DIVOTrack/blob/main/MOTChallengeEvalKit_cv_test/command.txt

Using command "python MOT/evalMOT.py", the result is as follows:

CVIDF1 and CVMA indicators are not included in the result.

How to get CVIDF1 and CVMA indicators?

As shown in https://github.com/shengyuhao/DIVOTrack/blob/main/MOTChallengeEvalKit_cv_test/command.txt

2. evaluate CVIDF1

MOT/evalMOT.py:

    gt_dir = "data/eval/wildscene_eval_cvma/gt"
    res_dir = "data/eval/wildscene_eval_cvma/track"
    seq_file = "data/eval/wildscene_eval_cvma/seqs.txt"

python MOT/evalMOT.py

3. evaluate CVMA

MOT/evalMOT.py:

    gt_dir = "data/eval/wildscene_eval_cvma/gt_cvma"
    res_dir = "data/eval/wildscene_eval_cvma/track_cvma"
    seq_file = "data/eval/wildscene_eval_cvma/seqs.txt"

python MOT/evalMOT.py

1. When you want to evaluate CVIDF1, change the file MOT/evalMOT.py to:

       gt_dir = "data/eval/{real_scene}_cvma/gt"
       res_dir = "data/eval/{real_scene}_cvma/track"
       seq_file = "data/eval/{real_scene}_cvma/seqs.txt"

   Run MOT/evalMOT.py, and the IDF1 column of the resulting table is the value of CVIDF1.

2. When you want to evaluate CVMA, change the file MOT/evalMOT.py to:

       gt_dir = "data/eval/{real_scene}_cvma/gt_cvma"
       res_dir = "data/eval/{real_scene}_cvma/track_cvma"
       seq_file = "data/eval/{real_scene}_cvma/seqs.txt"

   Run MOT/evalMOT.py, and the MOTA column of the resulting table is the value of CVMA.

[chen-si-jia]

Please refer to this file https://github.com/shengyuhao/DIVOTrack/blob/main/MOTChallengeEvalKit_cv_test/command.txt

Using command "python MOT/evalMOT.py", the result is as follows: CVIDF1 and CVMA indicators are not included in the result. How to get CVIDF1 and CVMA indicators?

As shown in https://github.com/shengyuhao/DIVOTrack/blob/main/MOTChallengeEvalKit_cv_test/command.txt

2. evaluate CVIDF1

MOT/evalMOT.py:

  gt_dir = "data/eval/wildscene_eval_cvma/gt"
  res_dir = "data/eval/wildscene_eval_cvma/track"
  seq_file = "data/eval/wildscene_eval_cvma/seqs.txt"

python MOT/evalMOT.py

3. evaluate CVMA

MOT/evalMOT.py:

  gt_dir = "data/eval/wildscene_eval_cvma/gt_cvma"
  res_dir = "data/eval/wildscene_eval_cvma/track_cvma"
  seq_file = "data/eval/wildscene_eval_cvma/seqs.txt"

python MOT/evalMOT.py

1. When you want to evaluate CVIDF1, change the file MOT/evalMOT.py to:

      gt_dir = "data/eval/{real_scene}_cvma/gt"
      res_dir = "data/eval/{real_scene}_cvma/track"
      seq_file = "data/eval/{real_scene}_cvma/seqs.txt"

   Run MOT/evalMOT.py, and the IDF1 column of the resulting table is the value of CVIDF1.

2. When you want to evaluate CVMA, change the file MOT/evalMOT.py to:

      gt_dir = "data/eval/{real_scene}_cvma/gt_cvma"
      res_dir = "data/eval/{real_scene}_cvma/track_cvma"
      seq_file = "data/eval/{real_scene}_cvma/seqs.txt"

   Run MOT/evalMOT.py, and the MOTA column of the resulting table is the value of CVMA.

Thank you!