Knight

About the pre-trained Re-ID model in DeepSORT.

1

Hi, I didn't find the link to store the pre-trained Re-ID [model,](https://github.com/ifzhang/ByteTrack/blob/9022caab72296ebf3a316a529101e9546968fd20/tools/track_deepsort.py#L110) does anyone know? Also what data was used to train this pre-trained model? Thanks.

Reproduced results on validation set MOT15 with the provided weights(motr_final_0915.pth), MOTA: 29.1% IDF1: 67.1%.

10

Thank you for your wondful work. I run the eval experiment on the MOT15 training set, the result as follows: KITTI-13 55.0% 43.8% 73.7% 74.2% 44.1% 42 23 18 1...

Hi, The length of the trajectory samples in the [training set](https://github.com/MedChaabane/DEFT/blob/71abba2c69a9d3a4d07154a1ed871f8846ba8648/src/lib/dataset/trajectory_dataset.py#L150) does not seem to be 10 as stated in the paper, but rather 9,10,11. Is this the reason why...

How can I get the supplementary material for this paper?

1

Hi, thank you for share your wonderful work! But how can I get the supplementary material for this paper?

How long before the AVA dataset is released? I'm very much looking forward to it.

Some questions.

8

I am very interested in your work. Can you provide the training code for the HMAR model and the annotation file for the AVA dataset? Thanks.

Is the preprocessed AVA dataset available now?

2

Hi, Is the preprocessed AVA dataset available now?

Thanks for your wonderful work! I'm wondering if there is a plan for releasing the code of 3DSL?

Thanks for open source this wonderful work, can you please share the script of figure 3 about motion offset statistics?

评测网站报错

8

评测网站出问题，提交后状态长时间显示为： processing, 最终显示error.即使提交此前能正常返回评估结果的csv文件也遇到这个错误。