Trainspotting: Learning Object-Centric 3D Reconstruction from Found Photos
cs.CV
Submitted: 2026-11-04
Updated: 2026-11-05
Terminology
Sources
- ShapeNet: An Information-Rich 3D Model Repository
- RoMa v2: Harder Better Faster Denser Feature Matching
- VGGT-
- Towards Consistent Video Geometry Estimation
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs