ArthurOuaknine

I have the same problem, did you find an easy way to keep the bounding boxes order ?

Thanks for your hint! Actually the output of the linear assignment function helps to build a mapping between the bounding box ids (order of the input) and the tracker ids...

Hello. Any updates about the Mask2Former integration? Thanks

Thanks for your work, it will definitely help :)

Thanks @Stakhan for your proposed solution. Note that running `pip install -r min_requirements.txt` raises `ERROR: No matching distribution found for open3d==1.13`. The pip install works by changing your `min_requirements.txt` with...

Hi, thanks for your interest in our work. Due to the some issues we encounter with the storage, the data is temporarily unavailable. We are currently working on a new...

Hello. In practice, the area under the precision-recall curve is simulated by computing the average of precisions for discretized values of recall (see your own wikipedia link, section "Average precision,...

Ok I see. It should be related to this loop: https://github.com/valeoai/RADIal/blob/ce58c8d114b0067e5b2eb4bca674ccd509444909/FFTRadNet/utils/metrics.py#L113 The current thresholds are `[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]`. The code should be `for threshold...

Thanks for your valuable comment, it is appreciated. I leave this issue open and let @jrebut do the integration and verify the performances.

Hello. Could you please provide an example? As far as I remember, the dataset should contain the same number of radar frames and camera images for each sequence.