Ref-DAVIS2017.md

May 27, 2023 ยท View on GitHub

Ref-DAVIS 2017

Model Zoo

As described in the paper, we report the results using the model trained on Ref-Youtube-VOS without finetune.

BackboneJ&FJFModel
ResNet-5065.362.468.2model
ResNet-10165.361.968.6model
Swin-L68.064.871.3model
Video-Swin-T66.563.070.0model
Video-Swin-S66.162.669.8model
Video-Swin-B66.462.870.0model
ConvNext-L69.065.672.4model
ConvMAE-B69.265.672.8model

Inference & Evaluation

./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