Question to authors: have you tested image augmentations in the training pipeline?

GoGoDuck912 / Self-Correction-Human-Parsing

An out-of-box human parsing representation extractor.

MIT License

1.03k stars 237 forks source link

Question to authors: have you tested image augmentations in the training pipeline? #60

Open FraPochetti opened 3 years ago

FraPochetti commented 3 years ago

Hi, first of all thanks a lot for the paper and the code. @PeikeLi I was wondering, have you tested using any image augmentations strategy in the training pipeline? Looking at your dataset, it doesn't seem like you are applying any. To be fair, LIP is large enough to probably make any augmentation redundant. Just curious though.