4 papers
GeoCache: Training-Free Acceleration of Multi-View Texture Diffusion via Geometric Delta Transport
Haotang Li, Zhenyu Qi, Shaohan Henry Wang +6
Geometry-conditioned multi-view diffusion enables high-quality 3D texture generation, but its repeated per-view denoiser evaluations introduce substantial computational cost. Exist…
PrePARE: Pre-AA Token Pruning for Frozen Multi-View Geometry Transformers
Haotang Li, Zhenyu Qi, Shaohan Henry Wang +5
Multi-view geometry transformers are feed-forward 3D foundation models that jointly predict depth maps, point maps, and camera poses for N images in a single forward pass. Most of…
MSSSeg: Learning Multi-Scale Structural Complexity for Self-Supervised Segmentation
Haotang Li, Zhenyu Qi, Hao Qin +4
Self-supervised semantic segmentation methods often suffer from structural errors, including merging distinct objects or fragmenting coherent regions, because they rely primarily o…
PhysDepth: Plug-and-Play Physical Refinement for Monocular Depth Estimation in Challenging Environments
Kebin Peng, Haotang Li, Zhenyu Qi +6
State-of-the-art monocular depth estimation (MDE) models often struggle in challenging environments, primarily because they overlook robust physical information. To demonstrate thi…