PromptStyler: Prompt-driven Style Generation for Source-free Domain Generalization

Adaptive Superpixel for Active Learning in Semantic Segmentation

Removing Multiple Biases through the Lens of Multi-task Learning

Stable and Consistent Prediction of 3D Characteristic Orientation via Invariant Residual Learning

Leveraging Proxy of Training Data for Test-Time Adaptation

Scaling up GANs for Text-to-Image Synthesis

Learning Rotation-Equivariant Features for Visual Correspondence

Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation

Devil's on the Edges: Selective Quad Attention for Scene Graph Generation

Human Pose Estimation in Extremely Low-Light Conditions