Research map / 3D vision / 3D generation
MV-GEL: Language-Driven Multi-View Geometric Entity Localization on Meshes
Kartik Bali, Roland Aydin
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 |
|---|---|---|---|
| fine-tune LISA on CAD renderings (LISA-CAD vs vanilla) (ablation) | MV-GEL FiLM selector · Face localization | IoU: 0.293 -> 0.501 | Improved |
| fine-tune LISA on CAD renderings (ablation) | MV-GEL FiLM selector · Edge localization | IoU: 0.046 -> 0.284 | Improved |
| use learned GELviews ranking instead of random views (ablation) | LISA-CAD · Face localization (top-1) | IoU: 0.196 -> 0.501 | Improved |
| use GELviews with view-interaction head instead of Only-CLIP (ablation) | LISA-CAD · Face localization (top-1) | IoU: 0.305 -> 0.501 | Improved |
| add modality fusion (FiLM) to view selector (ablation) | LISA-CAD · Face localization (top-1) | IoU: 0.490 -> 0.501 | Improved |
| use FiLM instead of Cross-Attention fusion (ablation) | LISA-CAD · Face localization (top-1) | IoU: 0.486 -> 0.501 | No clear effect |
| use Cross-Attention instead of FiLM fusion (ablation) | LISA-CAD · Edge localization (top-1) | IoU: 0.284 -> 0.287 | No clear effect |
| add view NMS to top-3 selection (ablation) | MV-GEL FiLM@top3 · Face localization | IoU: 0.419 -> 0.372 | Hurt |
| use top-1 instead of top-3 views (ablation) | MV-GEL FiLM · Face localization | IoU: IoU 0.419 -> 0.501; recall 0.810 -> 0.706 | Mixed |
| MV-GEL vs zero-shot point-cloud methods | MV-GEL FiLM@top1 · Face localization | IoU: 0.171 (Find3D) -> 0.501 | Improved |
Closest papers by meaning
- Let Language Constrain Geometry: Vision–Language Models as Semantic and Spatial Critics for 3D Generation (ICML 2026 · similarity 0.87)
- Beyond 3D VQAs: Injecting 3D Spatial Priors into Vision-Language Models for Enhanced Geometric Reasoning (CVPR 2026 · similarity 0.87)
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 2024 · similarity 0.87)
- GPVK-VL: Geometry-Preserving Virtual Keyframes for Visual Localization under Large Viewpoint Changes (CVPR 2025 · similarity 0.86)
- Segment Any 3D Object with Language (ICLR 2025 · similarity 0.86)
- Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering (NeurIPS 2026 · similarity 0.86)
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.
