1 paper · 1 filter
Zhiyang Lu, Ming Cheng
Cross-modal 2D-3D gait recognition is impeded by inherent domain discrepancies between 2D silhouette and 3D LiDAR range-view representations. While prior methods align only final e…