wjun0830 / QD-DETR

Official pytorch repository for "QD-DETR : Query-Dependent Video Representation for Moment Retrieval and Highlight Detection" (CVPR 2023 Paper)
https://arxiv.org/abs/2303.13874
Other
183 stars 13 forks source link

Eval #14

Closed Lonicer closed 1 year ago

Lonicer commented 1 year ago

Excuse me, CodaLab can only upload 5 times, how to evaluate the results of V+A?

wjun0830 commented 1 year ago

In our case, max submission was 100. You may want to ask the authors of MomentDETR