Seoul National University

Uncertainty-guided Compositional Alignment with Part-to-Whole Semantic Representativeness in Hyperbolic Vision-Language Models

Learning Human-Object Interaction for 3D Human Pose Estimation from LiDAR Point Clouds