Research map / 3D vision / Depth and stereo
On Geometric Understanding and Learned Priors in Feed-forward 3D Reconstruction Models
Jelena Bratulić, Sudhanshu Mittal, Thomas Brox, Christian Rupprecht
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| zero out high correspondence-matching attention heads in middle layers (ablation) | VGGT and DA3 · ShapeNet, 50mm focal length | Sampson distance error | Hurt |
| zero out random early or late layers' attention heads (ablation) | VGGT, DA3, DUSt3R · ShapeNet, 50mm focal length | Sampson distance error | No clear effect |
| more local knockout (patch level instead of whole head) (ablation) | VGGT, DA3, DUSt3R · ShapeNet | Sampson distance error degradation | Mixed |
| MLP probe vs linear probe for fundamental matrix recovery (ablation) | VGGT, DA3, DUSt3R · fundamental matrix probing | Sampson error | Improved |
| mask part of one image with white patches | VGGT · ShapeNet pairs, small viewpoint | Sampson error | No clear effect |
| change lighting on only one view | VGGT · ShapeNet lighting variations | failure rate | Hurt |
| add asymmetric lighting shadows to symmetric repeated scenes | VGGT · ShapeNet repetition scenes | geometric accuracy | Improved |
Closest papers by meaning
- Understanding Multi-View Transformers (NeurIPS 2026 · similarity 0.89)
- Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering (NeurIPS 2026 · similarity 0.88)
- Beyond 3D VQAs: Injecting 3D Spatial Priors into Vision-Language Models for Enhanced Geometric Reasoning (CVPR 2026 · similarity 0.88)
- Learning from Videos for 3D World: Enhancing MLLMs with 3D Vision Geometry Priors (NeurIPS 2025 · similarity 0.87)
- Probing the 3D Awareness of Visual Foundation Models (CVPR 2024 · similarity 0.87)
- Bridging Viewpoint Gaps: Geometric Reasoning Boosts Semantic Correspondence (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
