search

ZHKKKe / PixelSSL

A PyTorch-based Semi-Supervised Learning (SSL) Codebase for Pixel-wise (Pixel) Vision Tasks [ECCV 2020]
Apache License 2.0
289 stars 30 forks source link

Add implementation for Semi-supervised Semantic Segmentation via Strong-weak Dual-branch Network #5

Closed syorami closed 3 years ago

syorami commented 3 years ago

Thanks for your sharing and the repo is quite helpful for me to understand the work in SSL segmentation. If possible, could you add the implementation of Semi-supervised Semantic Segmentation via Strong-weak Dual-branch Network (ECCV 2020), which is a simply dual branch network. It's a quite easy and inituitive idea but I could not reproduce the results with deeplabv2. It would be great if you could add this into the repo.

ZHKKKe commented 3 years ago

Hi, thanks for your attention.

I will check it and response you ASAP.

ZHKKKe commented 3 years ago

@syorami I have checked the paper. The paper seems focus more on weak-supervised learning. I am sorry that we do not have enough time to reproduce it this month.

syorami commented 3 years ago

Does 'weak-supervised learning' indicate the usage of DSRG to generate weak annotations?

ZHKKKe commented 3 years ago

Oh... Yep... They use DSRG to generate weak annotations.. So we can regard the method as a semi-supervised method. Now we are reproducing S4GAN. We will consider reproducing this work then.

syorami commented 3 years ago

Got it. Thanks for your kindly response.