HYUN Jeongseok

@PeizeSun Isn't that Q and K must have the same hidden dimension to process matrix multiplication, like in DETR Q is [100 x C] and K is [HW x C]...

Hi. Sorry for the late reply. This logging output is not the error. When the visibility threshold of 0.0 is used, it means that all images should be used. Thus,...

That configuration trains SGT using half of the MOT17 training set and evaluates the other half. Actually, all experiments for the paper are conducted using the docker file with CUDA10.1....

Can you try these pretrained weights? [mot17half](https://hkustconnect-my.sharepoint.com/:u:/g/personal/jhyunaa_connect_ust_hk/EQPB4gGSEdpDpfeKNrXkAkQB6Gvail73wLgLIoo4i9z9Gg?e=iLvsJv) [mot17half+CH](https://hkustconnect-my.sharepoint.com/:u:/g/personal/jhyunaa_connect_ust_hk/EeYV09Ng6g5JmEsRYLs5IygB0Oh0MtBqJkRdRWaC-I55fg?e=DZWbtj)

It seems that you are using a different version of detectron2. Please install the detectron2 version stated in the dockerfile (read INSTALL.md of my repo). Otherwise, you should revise the...

https://github.com/sauradip/STALE/issues/23#issuecomment-1704056365 It seems yes.

Thank you for your reply! Q1. Then, when you obtained the results of THUMOS14 in a paper, have you used the filling operation? I really appreciate your clarification for me...

Thank you for your reply! Q3-1. The result of the MaskFormer code provided by you is a foreground logit score whose shape is B x nq x L [batch_size x...

Based on the available information, I can only achieve 2 mAP on THUMOS14 with the closed setting. The main bottleneck seems training action mask localizer part which is failed to...

Thanks for getting back to me! I'm a bit worried about the big differences between prediction and GT mask map. In each column, there seem to be too many points...