While fully-supervised deep learning yields good models for urban scene semantic segmentation, these models struggle to generalize to new environments with different lighting or weather conditions for instance. In addition, producing the extensive pixel-level annotations that the task requires comes at a great cost. Unsupervised domain adaptation (UDA) is one approach that tries to address these issues in order to make such systems more scalable. In particular, self-supervised learning (SSL) has recently become an effective strategy for UDA in semantic segmentation. At the core of such methods lies `pseudo-labeling', that is, the practice of assigning high-confident class predictions as pseudo-labels, subsequently used as true labels, for target data. To collect pseudo-labels, previous works often rely on the highest softmax score, which we here argue as an unfavorable confidence measurement. In this work, we propose Entropy-guided Self-supervised Learning (ESL), leveraging entropy as the confidence indicator for producing more accurate pseudo-labels. On different UDA benchmarks, ESL consistently outperforms strong SSL baselines and achieves state-of-the-art results.
@article{arxiv.2006.08658,
title = {ESL: Entropy-guided Self-supervised Learning for Domain Adaptation in Semantic Segmentation},
author = {Antoine Saporta and Tuan-Hung Vu and Matthieu Cord and Patrick Pérez},
journal= {arXiv preprint arXiv:2006.08658},
year = {2020}
}
Comments
Accepted at the CVPR 2020 Workshop on Scalability in Autonomous Driving