Segment scene in new domain unsupervised by refine pseudo labels and predict noisy labels with PRN
Segment scene in new domain unsupervised by refine pseudo labels and predict noisy labels with PRN
Unsupervised Domain Adaptation for Semantic Segmentation with Pseudo Label Self-Refinement
arXiv paper abstract https://arxiv.org/abs/2310.16979
arXiv PDF paper https://arxiv.org/pdf/2310.16979.pdf
Deep learning ... for semantic segmentation suffer ... degradation when tested on data with different characteristics than ... during the training.
... Unsupervised Domain Adaptation (UDA) approaches are crucial in deploying these models in the actual operating conditions.
Recent ... methods employ ... self-training approach, where a teacher model is used to generate pseudo-labels for the new data which in turn guide the training process of the student model.
Though this approach has seen a lot of success, it suffers from the issue of noisy pseudo-labels being propagated in the training process.
... propose an auxiliary pseudo-label refinement network (PRN) for online refining of the pseudo labels and also localizing the pixels whose predicted labels are likely to be noisy. ... improve the quality of pseudo labels and select highly reliable ones
... approach ... performs significantly better than the previous state-of-the-art methods.
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comentarios