2 papers
cs.CV2026
Residual Flow Matching with Dynamic Cross-Interaction for 3D Multi-Person Motion Prediction
Wei Wei, Yinyuan Zhao, Ruixuan Yu
3D multi-person motion prediction requires modeling both individual kinematics and inter-person interactions. While Flow Matching is effective for multi-hypothesis generation to im…
cs.CV2025
UniLat3D: Geometry-Appearance Unified Latents for Single-Stage 3D Generation
Guanjun Wu, Jiemin Fang, Chen Yang +11
High-fidelity 3D asset generation is crucial for various industries. While recent 3D pretrained models show strong capability in producing realistic content, most are built upon di…