Self-supervised Augmentation Consistency for Adapting Semantic Segmentation
This work addresses the problem of adapting semantic segmentation models to new domains for practitioners, offering a simpler and more effective solution compared to previous approaches.
The paper tackles domain adaptation for semantic segmentation by using self-supervised augmentation consistency instead of complex adversarial methods, achieving significant improvements in state-of-the-art segmentation accuracy across various architectures and scenarios.
We propose an approach to domain adaptation for semantic segmentation that is both practical and highly accurate. In contrast to previous work, we abandon the use of computationally involved adversarial objectives, network ensembles and style transfer. Instead, we employ standard data augmentation techniques $-$ photometric noise, flipping and scaling $-$ and ensure consistency of the semantic predictions across these image transformations. We develop this principle in a lightweight self-supervised framework trained on co-evolving pseudo labels without the need for cumbersome extra training rounds. Simple in training from a practitioner's standpoint, our approach is remarkably effective. We achieve significant improvements of the state-of-the-art segmentation accuracy after adaptation, consistent both across different choices of the backbone architecture and adaptation scenarios.