search

PeterL1n / RobustVideoMatting

Robust Video Matting in PyTorch, TensorFlow, TensorFlow.js, ONNX, CoreML!
https://peterl1n.github.io/RobustVideoMatting/
GNU General Public License v3.0
8.61k stars 1.14k forks source link

Why is youtubevis seg resized with bilinear? #119

Closed haichaoyu closed 2 years ago

haichaoyu commented 2 years ago

Hi Peter,

Thanks for the great repo. One question regarding the code: why is the youtubevis labels resized with bilinear mode? Should it be nearest as the other datasets?

https://github.com/PeterL1n/RobustVideoMatting/blob/master/dataset/youtubevis.py#L109

Thanks.

PeterL1n commented 2 years ago

I guess you can change it to NEAREST. Shouldn't matter too much.