Multi-source Domain Adaptation for Semantic Segmentation

Abstract

Simulation-to-real domain adaptation for semantic segmentation has been actively studied for various applications such as autonomous driving. Existing methods mainly focus on a single-source setting, which cannot easily handle a more practical scenario of multiple sources with different distributions. In this paper, we propose to investigate multi-source domain adaptation for semantic segmentation. Specifically, we design a novel framework, termed Multi-source Adversarial Domain Aggregation Network (MADAN), which can be trained in an end-to-end manner. First, we generate an adapted domain for each source with dynamic semantic consistency while aligning at the pixel-level cycle-consistently towards the target. Second, we propose sub-domain aggregation discriminator and cross-domain cycle discriminator to make different adapted domains more closely aggregated. Finally, feature-level alignment is performed between the aggregated domain and target domain while training the segmentation network. Extensive experiments from synthetic GTA and SYNTHIA to real Cityscapes and BDDS datasets demonstrate that the proposed MADAN model outperforms state-of-the-art approaches. Our source code is released at: https://github.com/Luodian/MADAN.

Accepted by NeurIPS 2019

Cited by in corpus (26)

• Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation
• Uncertainty-Aware Consistency Regularization for Cross-Domain Semantic Segmentation
• DAugNet: Unsupervised, Multi-source, Multi-target, and Life-long Domain Adaptation for Semantic Segmentation of Satellite Images
• Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
• Multi-source Domain Adaptation in the Deep Learning Era: A Systematic Survey
• Rethinking Distributional Matching Based Domain Adaptation
• Multi-site fMRI Analysis Using Privacy-preserving Federated Learning and Domain Adaptation: ABIDE Results
• Multi-Target Domain Adaptation with Collaborative Consistency Learning
• Multi-source Pseudo-label Learning of Semantic Segmentation for the Scene Recognition of Agricultural Mobile Robots
• Co-Training for Unsupervised Domain Adaptation of Semantic Segmentation Models
• Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentation
• Learning Invariant Representations and Risks for Semi-supervised Domain Adaptation
• VersatileGait: A Large-Scale Synthetic Gait Dataset with Fine-GrainedAttributes and Complicated Scenarios
• SF-UDA${}^{3D}$: Source-Free Unsupervised Domain Adaptation for LiDAR-Based 3D Object Detection
• A Theory of Label Propagation for Subpopulation Shift
• Multi-source Distilling Domain Adaptation
• MADAN: Multi-source Adversarial Domain Aggregation Network for Domain Adaptation
• Union-set Multi-source Model Adaptation for Semantic Segmentation
• Multi-source Domain Adaptation for Visual Sentiment Classification
• Multi-source Few-shot Domain Adaptation
• Curriculum CycleGAN for Textual Sentiment Domain Adaptation with Multiple Sources
• Discriminative Noise Robust Sparse Orthogonal Label Regression-based Domain Adaptation
• Cross-Domain Structure Preserving Projection for Heterogeneous Domain Adaptation
• EvDistill: Asynchronous Events to End-task Learning via Bidirectional Reconstruction-guided Cross-modal Knowledge Distillation
• VersatileGait: A Large-Scale Synthetic Gait Dataset Towards in-the-Wild Simulation
• Densely Semantic Enhancement for Domain Adaptive Region-free Detectors