Category Contrast for Unsupervised Domain Adaptation in Visual Tasks
arXiv:2106.02885
Abstract
Instance contrast for unsupervised representation learning has achieved great success in recent years. In this work, we explore the idea of instance contrastive learning in unsupervised domain adaptation (UDA) and propose a novel Category Contrast technique (CaCo) that introduces semantic priors on top of instance discrimination for visual UDA tasks. By considering instance contrastive learning as a dictionary look-up operation, we construct a semantics-aware dictionary with samples from both source and target domains where each target sample is assigned a (pseudo) category label based on the category priors of source samples. This allows category contrastive learning (between target queries and the category-level dictionary) for category-discriminative yet domain-invariant feature representations: samples of the same category (from either source or target domain) are pulled closer while those of different categories are pushed apart simultaneously. Extensive UDA experiments in multiple visual tasks (e.g., segmentation, classification and detection) show that CaCo achieves superior performance as compared with state-of-the-art methods. The experiments also demonstrate that CaCo is complementary to existing UDA methods and generalizable to other learning setups such as unsupervised model adaptation, open-/partial-set adaptation etc.
CVPR2022 version
References in corpus (12)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Learning Transferable Features with Deep Adaptation Networks
- Improved Baselines with Momentum Contrastive Learning
- FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
- Learning Representations by Maximizing Mutual Information Across Views
- A Theoretical Analysis of Contrastive Unsupervised Representation Learning
- Supervised Contrastive Learning
- Supervised Contrastive Learning for Pre-trained Language Model Fine-tuning
- Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation
- FSDR: Frequency Space Domain Randomization for Domain Generalization
- MLAN: Multi-Level Adversarial Network for Domain Adaptive Semantic Segmentation
- Cross-View Regularization for Domain Adaptive Panoptic Segmentation