search

hkchengrex / Mask-Propagation

[CVPR 2021] MiVOS - Mask Propagation module. Reproduced STM (and better) with training code :star2:. Semi-supervised video object segmentation evaluation.
https://hkchengrex.github.io/MiVOS/
MIT License
128 stars 22 forks source link

metrics results of test dataset #40

Closed xwhkkk closed 2 years ago

xwhkkk commented 2 years ago

After I run the code _eval_davis2016.py, I only get the mask file in the output file. how could I get the value of metrics such as J, J&F? and how could we test the model on personal datasets to get those metrics after using _interactivegui.py? Thanks for your suggestions

hkchengrex commented 2 years ago
  1. Use DAVIS's official tool: https://davischallenge.org/
  2. You cannot unless you have the ground truths.
xwhkkk commented 2 years ago

Yes, I have the ground truths. Is it means I need to write my own code to evaluate the metrics or I can referee the DAVIS official tool?

hkchengrex commented 2 years ago

You can map your data into DAVIS format and use DAVIS's tool.

xwhkkk commented 2 years ago

Thanks for your reply! It works well.