As described in the paper, we report the results using the model trained on Ref-Youtube-VOS without finetune.
| Backbone | J&F | J | F | Model |
|---|---|---|---|---|
| ResNet-50 | 65.3 | 62.4 | 68.2 | model |
| ResNet-101 | 65.3 | 61.9 | 68.6 | model |
| Swin-L | 68.0 | 64.8 | 71.3 | model |
| Video-Swin-T | 66.5 | 63.0 | 70.0 | model |
| Video-Swin-S | 66.1 | 62.6 | 69.8 | model |
| Video-Swin-B | 66.4 | 62.8 | 70.0 | model |
| ConvNext-L | 69.0 | 65.6 | 72.4 | model |
| ConvMAE-B | 69.2 | 65.6 | 72.8 | model |
./scripts/dist_test_davis.sh --backbone [backbone]
For example, evaluating the Swin-Large model, run the following command:
./scripts/dist_test_davis.sh --backbone swin_l_p4w7
Note that, if you use the weights we provide, you should put the weights in the corresponding path. ./results/[backbone]/ckpt/backbone_weight.pth