Adversarial Style Mining for One-Shot Unsupervised Domain Adaptation

Abstract

We aim at the problem named One-Shot Unsupervised Domain Adaptation. Unlike traditional Unsupervised Domain Adaptation, it assumes that only one unlabeled target sample can be available when learning to adapt. This setting is realistic but more challenging, in which conventional adaptation approaches are prone to failure due to the scarce of unlabeled target data. To this end, we propose a novel Adversarial Style Mining approach, which combines the style transfer module and task-specific module into an adversarial manner. Specifically, the style transfer module iteratively searches for harder stylized images around the one-shot target sample according to the current learning state, leading the task model to explore the potential styles that are difficult to solve in the almost unseen target domain, thus boosting the adaptation performance in a data-scarce scenario. The adversarial learning framework makes the style transfer module and task-specific module benefit each other during the competition. Extensive experiments on both cross-domain classification and segmentation benchmarks verify that ASM achieves state-of-the-art adaptation performance under the challenging one-shot setting.

Preprint

References in corpus (10)

• ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation
• Conditional Adversarial Domain Adaptation
• The Cityscapes Dataset for Semantic Urban Scene Understanding
• FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
• Adversarial Discriminative Domain Adaptation
• One-Shot Unsupervised Cross Domain Translation
• Domain Stylization: A Strong, Simple Baseline for Synthetic to Real Image Domain Adaptation
• Semantic-aware Grad-GAN for Virtual-to-Real Urban Scene Adaption
• Fully Convolutional Adaptation Networks for Semantic Segmentation
• Significance-aware Information Bottleneck for Domain Adaptive Semantic Segmentation

Cited by in corpus (19)

• A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts
• Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation
• Free-form tumor synthesis in computed tomography images via richer generative adversarial network
• Adversarial Domain Adaptation with Prototype-Based Normalized Output Conditioner
• Inter-Domain Mixup for Semi-Supervised Domain Adaptation
• Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation
• MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaption
• Exploring UMAP in hybrid models of entropy-based and representativeness sampling for active learning in biomedical segmentation
• Few-shot Unsupervised Domain Adaptation with Image-to-class Sparse Similarity Encoding
• Unsupervised Eyeglasses Removal in the Wild
• Encoding Robustness to Image Style via Adversarial Feature Perturbations
• Federated Generative Adversarial Learning
• Automated Deepfake Detection
• Omni-supervised Facial Expression Recognition via Distilled Data
• One-shot Unsupervised Domain Adaptation with Personalized Diffusion Models
• Point Adversarial Self Mining: A Simple Method for Facial Expression Recognition
• Look, Cast and Mold: Learning 3D Shape Manifold from Single-view Synthetic Data
• PØDA: Prompt-driven Zero-shot Domain Adaptation
• Improving Weakly-supervised Object Localization via Causal Intervention