wangxiang1230 / SSTAP

Code for our CVPR 2021 Paper "Self-Supervised Learning for Semi-Supervised Temporal Action Proposal".
70 stars 8 forks source link

How can I print mAP values instead of AR@AN #13

Open 0829MingHu opened 1 year ago

0829MingHu commented 1 year ago

Thank you so much for your excellent work. I am reproducing your model on my own dataset, I ask how can I print mAP values instead of AR@AN?

wangxiang1230 commented 1 year ago

Thank you so much for your excellent work. I am reproducing your model on my own dataset, I ask how can I print mAP values instead of AR@AN?

Hi, thanks for your attention to SSTAP. For ActivityNet 1.3, the detection code is here. We follow the "proposal + classification" paradigm to obtain the detection results. For your own dataset, I suggest you directly use the features of the proposal after ROI pooling for proposal classification.