Haochen-Wang409/U2PL
PyTorch implementation of semi-supervised semantic segmentation that leverages unreliable pseudo-labels to improve training with limited labeled data.
Not currently ranked — collecting fresh signals.
star history
This repository provides an official implementation of a research paper on semi-supervised semantic segmentation. The method addresses the problem of underutilized pixels in unlabeled images by assigning pseudo-labels to both reliable and unreliable predictions, expanding the training data effectively. It uses deep learning with PyTorch and includes benchmarks on Cityscapes and PASCAL VOC datasets.
Frequently asked
- What is Haochen-Wang409/U2PL?
- PyTorch implementation of semi-supervised semantic segmentation that leverages unreliable pseudo-labels to improve training with limited labeled data.
- Is U2PL open source?
- Yes — Haochen-Wang409/U2PL is open source, released under the Apache-2.0 license.
- What language is U2PL written in?
- Haochen-Wang409/U2PL is primarily written in Python.
- How popular is U2PL?
- Haochen-Wang409/U2PL has 480 stars on GitHub.
- Where can I find U2PL?
- Haochen-Wang409/U2PL is on GitHub at https://github.com/Haochen-Wang409/U2PL.