DSP: Dual Soft-Paste for Unsupervised Domain Adaptive Semantic Segmentation

Abstract

Unsupervised domain adaptation (UDA) for semantic segmentation aims to adapt a segmentation model trained on the labeled source domain to the unlabeled target domain. Existing methods try to learn domain invariant features while suffering from large domain gaps that make it difficult to correctly align discrepant features, especially in the initial training phase. To address this issue, we propose a novel Dual Soft-Paste (DSP) method in this paper. Specifically, DSP selects some classes from a source domain image using a long-tail class first sampling strategy and softly pastes the corresponding image patch on both the source and target training images with a fusion weight. Technically, we adopt the mean teacher framework for domain adaptation, where the pasted source and target images go through the student network while the original target image goes through the teacher network. Output-level alignment is carried out by aligning the probability maps of the target fused image from both networks using a weighted cross-entropy loss. In addition, feature-level alignment is carried out by aligning the feature maps of the source and target images from student network using a weighted maximum mean discrepancy loss. DSP facilitates the model learning domain-invariant features from the intermediate domains, leading to faster convergence and better performance. Experiments on two challenging benchmarks demonstrate the superiority of DSP over state-of-the-art methods. Code is available at \url{https://github.com/GaoLii/DSP}.

References in corpus (18)

• mixup: Beyond Empirical Risk Minimization
• Generative Adversarial Networks
• Learning Transferable Features with Deep Adaptation Networks
• Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
• The Cityscapes Dataset for Semantic Urban Scene Understanding
• FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
• CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
• Progressive LiDAR Adaptation for Road Detection
• ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias
• Playing for Data: Ground Truth from Computer Games
• FMix: Enhancing Mixed Sample Data Augmentation
• Category Anchor-Guided Unsupervised Domain Adaptation for Semantic Segmentation
• Uncertainty-Aware Consistency Regularization for Cross-Domain Semantic Segmentation
• Affinity Space Adaptation for Semantic Segmentation Across Domains
• Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation
• Self-Ensembling with GAN-based Data Augmentation for Domain Adaptation in Semantic Segmentation
• MetaCorrection: Domain-aware Meta Loss Correction for Unsupervised Domain Adaptation in Semantic Segmentation
• Unsupervised Domain Adaptation for Semantic Segmentation by Content Transfer

Cited by in corpus (8)

• An Empirical Study of Remote Sensing Pretraining
• Learning to Purification for Unsupervised Person Re-identification
• All for One, and One for All: UrbanSyn Dataset, the third Musketeer of Synthetic Driving Scenes
• Co-Training for Unsupervised Domain Adaptation of Semantic Segmentation Models
• Both Style and Fog Matter: Cumulative Domain Adaptation for Semantic Foggy Scene Understanding
• Unified Domain Adaptive Semantic Segmentation
• Pseudolabel guided pixels contrast for domain adaptive semantic segmentation
• ECAP: Extensive Cut-and-Paste Augmentation for Unsupervised Domain Adaptive Semantic Segmentation