search

chengche6230 / ReST

[ICCV 2023] ReST: A Reconfigurable Spatial-Temporal Graph Model for Multi-Camera Multi-Object Tracking
MIT License
130 stars 15 forks source link

Detection part #7

Closed Ququqq closed 5 months ago

Ququqq commented 7 months ago

Hello, thank you very much for sharing the code. I am a newbie and recently wanted to try running your code and had a lot of problems along the way. I'd appreciate it if you could help me with my doubts.

  1. Question1: Prepare all image frames as {FRAME}_{CAM}.jpg in /output/frames.
    This line of commentary of yours, doesn't it distinguish which camera is shooting the footage? For example, there are two cameras with a total of 200 frames. The file name is eg.1_CAM --200_CAM? Can you look at the image form under your folder.
  2. Question2: Select what input detection you want, and modify MODEL.DETECTION Where do I download the testing section Looking forward to your reply. Can't wait to contact you ，my email：ququu0201@163.com thanks
chengche6230 commented 7 months ago

Hi,

Sorry for not getting back to you sooner because of the day-off.

Question 1

The name of image frames will be like:

Question 2

For groundtruth detection, you can access it via directly modifying MODEL.DETECTION to 'gt'. As for the custom detection, you need to prepare the detection file following the same format as gt_test.json first.