Official Pytorch implementation of "Learnable Gated Temporal Shift Module for Deep Video Inpainting. Chang et al. BMVC 2019." and the FVI dataset in "Free-form Video Inpainting with 3D Gated Convolution and Temporal PatchGAN, Chang et al. ICCV 2019"
The issue #23 mentioned that an argument -mask of evaluate.py described in doc/script_usage.md does not exist in evaluate.py. After investigation, I found this argument is removed in the old repository because all evaluation should be done along with masks:
So, I remove -mask in doc/script_usage.md to prevent future confusion.
PR readiness check list
[ v ] Did it pass the Flake8 check?
[ v ] Did you test this PR?
Nash2325138
commented
4 years ago 
Why do we need this PR?
The issue #23 mentioned that an argument
-maskofevaluate.pydescribed indoc/script_usage.mddoes not exist inevaluate.py. After investigation, I found this argument is removed in the old repository because all evaluation should be done along with masks:So, I remove
-maskindoc/script_usage.mdto prevent future confusion.PR readiness check list
[ v ] Did it pass the Flake8 check? [ v ] Did you test this PR?