VIRD: View-Invariant Representation through Dual-Axis Transformation for Cross-View Pose Estimation figure
AlphaXiv 中文概览(可滚动查看)