Awesome
RPL
<img src="https://github.com/yyliu01/RPL/assets/102338056/4307ce17-9c44-4e19-82b7-b0508f51ff28.png" width="700" height="300" />Residual Pattern Learning for Pixel-wise Out-of-Distribution Detection in Semantic Segmentation
by Yuyuan Liu*, Choubo Ding*, Yu Tian, Guansong Pang, Vasileios Belagiannis, Ian Reid and Gustavo Carneiro
Installation
please install the dependencies and dataset based on this installation document.
Getting start
please follow this instruction document to reproduce our results.
Results
our training logs and checkpoints are in this result page.
Acknowledgement & Citation
Our code is highly based on the PEBAL. Please consider citing them in your publications if they help your research.
@inproceedings{liu2023residual,
title={Residual pattern learning for pixel-wise out-of-distribution detection in semantic segmentation},
author={Liu, Yuyuan and Ding, Choubo and Tian, Yu and Pang, Guansong and Belagiannis, Vasileios and Reid, Ian and Carneiro, Gustavo},
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
pages={1151--1161},
year={2023}
}
@inproceedings{tian2022pixel,
title={Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes},
author={Tian, Yu and Liu, Yuyuan and Pang, Guansong and Liu, Fengbei and Chen, Yuanhong and Carneiro, Gustavo},
booktitle={European Conference on Computer Vision},
pages={246--263},
year={2022},
organization={Springer}
}
TODO
- RPL code has been released.
- RPL+CoroCL code has been released.
- The results based on extra training sets (e.g., Vistas, Wilddash2) have been released.