MCG-NJU/VideoMAE-Action-Detection

Issues

• Could you provide AVA2.2 finetuned weights for ViT-L and ViT-H?

  #6 opened 6 months ago by payne4handsome
  0

• How can I use pre-trained model to detect action directly?

  #5 opened a year ago by vba34520
  0

• What's the "extra label" setting in AVA 2.2 results

  #2 opened a year ago by ylz0925
  1

• How to test?

  #4 opened 2 years ago by Seeseallllll
  0

• Input resolution of train and validation

  #3 opened 2 years ago by ylz0925
  0

• AVA model checkpoint for ViT-L & ViT-H ?

  #1 opened 2 years ago by lisosia
  0