Search papers, labs, and topics across Lattice.
KAIST
2
0
5
LLMs still struggle to effectively leverage spatial and temporal reasoning tools for multi-camera person search, even with explicit graph representations of camera networks and transition times.
By adapting diffusion features in 3D Gaussian space, GeoNVS achieves state-of-the-art novel view synthesis with significantly improved geometric fidelity and camera control compared to existing video diffusion models.
