Research map / 3D vision / Face and shape
3D Highlighter: Localizing Regions on 3D Shapes via Text Descriptions
Dale Decatur, Itai Lang, Rana Hanocka
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 |
|---|---|---|---|
| Use 3D Highlighter instead of LSeg | text-driven 3D mesh highlighting · Highlight fidelity | CLIP R-Precision (ViT-L/14): 18.75 -> 81.25 | Improved |
| Use 3D Highlighter instead of Text2LIVE | text-driven 3D mesh highlighting · Highlight fidelity | CLIP R-Precision (ViT-L/14): 43.75 -> 81.25 | Improved |
| Use 3D Highlighter instead of LSeg | text-driven 3D mesh highlighting · Highlight fidelity | CLIP R-Precision (ViT-B/16): 6.25 -> 43.75 | Improved |
| Use 3D Highlighter instead of Text2LIVE | text-driven 3D mesh highlighting · Highlight fidelity | CLIP R-Precision (ViT-B/16): 31.25 -> 43.75 | Improved |
| Use a neural field rather than optimizing vertex colors directly (ablation) | 3D Highlighter · Text-driven mesh highlighting | highlight coherence | Improved |
| Add probability-weighted color blending (ablation) | 3D Highlighter · Text-driven mesh highlighting | highlight coherence | Improved |
| Add 2D perspective augmentations (ablation) | 3D Highlighter · Text-driven mesh highlighting | localization quality | Improved |
| Omit positional encoding (ablation) | 3D Highlighter neural field · Text-driven mesh highlighting | highlight coherence | Improved |
| Initialize probabilities near 0.5 rather than 0 (ablation) | 3D Highlighter · Text-driven mesh highlighting | localization quality | Improved |
| Initialize probabilities near 0.5 rather than 1 (ablation) | 3D Highlighter · Text-driven mesh highlighting | localization quality | Improved |
| Choose a CLIP-selected primary view and sample nearby views rather than always including the anchor view (ablation) | 3D Highlighter · Text-driven mesh highlighting | localization quality | Improved |
| Center view sampling on a recognizable primary view instead of sampling uniformly (ablation) | 3D Highlighter · Text-driven mesh highlighting | localization quality | Improved |
Later papers that compared with this method
2 later accepted papers report 3D Highlighter as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- 3D Paintbrush: Local Stylization of 3D Shapes with Cascaded Score Distillation (CVPR 2024 · better on 1 of 1 shared result)
- MeshSegmenter: Zero-Shot Mesh Segmentation via Texture Synthesis (ECCV 2024 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multi3DRefer: Grounding Text Description to Multiple 3D Objects (ICCV 2023 · similarity 0.86)
- ScanRefer: 3D Object Localization in RGB-D Scans using Natural Language (ECCV 2020 · similarity 0.86)
- Progressive3D: Progressively Local Editing for Text-to-3D Content Creation with Complex Semantic Prompts (ICLR 2024 · similarity 0.85)
- 3D Space as a Scratchpad for Editable Text-to-Image Generation (CVPR 2026 · similarity 0.84)
- From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes (NeurIPS 2025 · similarity 0.84)
- Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding (CVPR 2025 · similarity 0.84)
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.
