Code used for the results in the paper A Pseudo-Dual Self-Rectification Framework for Semantic Segmentation Abstract.Semantic segmentation has achieved remarkable success in various applications. However, the training process for such techniques necessitates a significant amount of labeled data. Although semi-supervised frameworks can alleviate this issue, traditional approaches typically require multiple baseline models to form a dual model. To allow a semi-supervised semantic segmentation framework to be used in robotic systems with precious computation and memory resources, we propose a framework utilizing a single baseline model only. The overall framework is composed of three parts: an encoder, a shallow decoder, and a deep decoder. It distills knowledge from the ensemble of two decoders to improve the encoder, which can implicitly form a pseudo-dual model. It also calculates classwise likelihoods according to the similarity between features and class prototypes learned from different decoders and rectifies low-confidence pseudo-labels. Our framework outperforms stateof-the-art frameworks on benchmark datasets with a significant amount of decrease in using computing resources.
mkdir ../data/CityScapes/
Download the dataset from here.
mkdir ../data/voc/
Download the dataset from here.
python train_kl.py --snapshot-dir ./result/2/kl --drop 0.1 --batch-size 2 --learning-rate 1e-4 --crop-size 256,512
python train_pro.py -snapshot-dir ./result/2/pro --drop 0.1 --batch-size 2 --learning-rate 1e-4 --crop-size 256,512
@inproceedings{hao2023pseudo, title={A Pseudo-Dual Self-Rectification Framework for Semantic Segmentation}, author={Hao, Huazheng and Xiao, Hui and Dong, Li and Yan, Diqun and Liang, Dongtai and Zhuang, Jiayan and Peng, Chengbin}, booktitle={2023 IEEE International Conference on Multimedia and Expo (ICME)}, pages={408--413}, year={2023}, organization={IEEE} }