OpenGVLab/ego4d-eccv2022-solutions

Issues

• ego4d_verb_pretrain_vitl_k700.pt Top-1: 16.60%, Top-5: 49.78%

  #12 opened 10 months ago by elmsamcht2189
  6

• Question about the provided video feature for MQ and NLQ

  #11 opened a year ago by yingsen1
  1

• STA pretrained model checkpoint

  #9 opened a year ago by kim-sanghwan
  1

• code for NLQ fusion

  #10 opened a year ago by Yisen-Feng
  2

• How to modify number of temporal views for the FHP task

  #8 opened a year ago by kurshakuz
  2

• SCOD How to visualize bounding boxes of model?

  #3 opened a year ago by Davidyao99
  1

• Access to pretrained model for STA task

  #7 opened a year ago by kurshakuz
  0

• Access to the FHP checkpoints

  #5 opened a year ago by kurshakuz
  4

• Hardware Requirements for Training?

  #6 opened a year ago by bryanzhou008
  1

• Would you release the leading model and code for SCOD with Swin-L and Pre-trained on ImageNet-22K + Objects365

  #4 opened 2 years ago by bryanzhou008
  6

• Validation data for SCOD

  #2 opened 2 years ago by Davidyao99
  5

• NLQ(*) and MQ(*) backbone checkpoints

  #1 opened 2 years ago by srama2512
  12