v0.9.0

Highlights

• Support Single-lens MOT (#258)
• Support video recognition (X3D, SWIN-video) (#256)

New Features

• Add inception config and voc config for FCN and UperNet (#261)
• Add inference time under V100 for the benchmark of deitiii and hydra attention (#251)
• Add bev-blancehybrid benchmark (#249)

Improvements

• Optimize data source apis (#254)
• Update predict.py to support input model directory (#252)

Bug Fixes

• Fix MAE arg error after timm upgrade (#255)
• Fix export SSL models bug, avoid loading default pretrained backbone model (#257)
• Fix bug can't find config files while easycv is installed (#253)