search

hkchengrex / MiVOS

[CVPR 2021] Modular Interactive Video Object Segmentation: Interaction-to-Mask, Propagation and Difference-Aware Fusion. Semi-supervised VOS as well!
https://hkchengrex.com/MiVOS/
MIT License
469 stars 64 forks source link

static dataset in download_dataset.py #41

Closed nazimii closed 2 years ago

nazimii commented 2 years ago

I note that there are a static dataset in download_dataset.py so, where is this static dataset used?

and in readme.md, you say, you use BL30K to train fusion model, and the BL30K is very large(600G), so ,you use 600 G dataset to pretrain fusion model?

hkchengrex commented 2 years ago
  1. For pretraining. See our paper or the STM paper.
  2. A subset of it. You can download our pre-generated fusion_data to check but I don't think it matters too much. The fusion module is pretty easy to train.
nazimii commented 2 years ago

thanks.. so, you use BL30K_a 0~1000 video to pretrain fusion model?

hkchengrex commented 2 years ago

I couldn't recall but I used fusion_data. I don't have that on my computer right now but if that's what you saw it should be correct.