Neutral Cross-Entropy Loss Based Unsupervised Domain Adaptation for Semantic Segmentation
- 20 April 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 4516-4525
Abstract
The generalization performance for semantic segmentation remains a major challenge when the data distributions between the source and target domain mismatch. Unsupervised domain adaptation (UDA) approaches are proposed to mitigate the problem above, among which entropy-minimization-based methods have gained more and more attention. However, the methods merely follow the cluster assumption sharpening the prediction distribution, thus have limited performance improvement. Without additional priors, the entropy loss can easily over-sharpen the prediction distribution, which brings noisy information into the learning process. On the other hand, the gradient of the entropy loss is strongly biased toward easy samples, also leading to limited generalization advances. In this paper, we firstly propose a pixel-level consistency regularization method, which introduces the smoothness prior to the UDA problem. Furthermore, we propose the neutral cross-entropy loss based on the consistency regularization, and reveal that its internal neutralization mechanism mitigates the over-sharpness of entropy minimization via the flatness effect of consistency regularization. We also demonstrate that the gradient bias toward easy samples is inherently tackled via the neutral cross-entropy loss. The experiments show that the proposed method has outperformed state-of-the-art methods in two synthetic-to-real experiments, only using the lightweight network.Keywords
Funding Information
- National Natural Science Foundation of China (U1764264/61873165)
This publication has 25 references indexed in Scilit:
- ERFNet: Efficient Residual Factorized ConvNet for Real-Time Semantic SegmentationIEEE Transactions on Intelligent Transportation Systems, 2017
- Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks2017
- Pyramid Scene Parsing Network2017
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
- Playing for Data: Ground Truth from Computer Games2016
- The SYNTHIA Dataset: A Large Collection of Synthetic Images for Semantic Segmentation of Urban Scenes2016
- Deep Residual Learning for Image Recognition2016
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016
- Fully convolutional networks for semantic segmentation2015
- The Pascal Visual Object Classes Challenge: A RetrospectiveInternational Journal of Computer Vision, 2014