Research map / 3D vision / 3D generation
PARIS3D: Reasoning-based 3D Part Segmentation Using Large Multimodal Model
Amrin Kareem, Jean Lahoud, Hisham Cholakkal
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 multi-view on RPSeg3D part data | LISA-MV · RPSeg3D Test overall | category mIoU: 19.08 -> 50.78 | Improved |
| fine-tune with 3D queries and explanations (PARIS3D) vs fine-tuned LISA-MV | PARIS3D · RPSeg3D Test overall | category mIoU: 50.78 -> 56.77 | Improved |
| add colour information to GPT-generated training instructions (ablation) | PARIS3D · 5 part categories | category mIoU: 46.7 -> 50.8 | Improved |
| add 3D information (position, shape, dimensions) to training instructions (ablation) | PARIS3D · 5 part categories | category mIoU: 50.8 -> 57.6 | Improved |
| increase rendered views from 1 to 5 to 10 (ablation) | PARIS3D · Bottle | category mIoU: 1.1 -> 23.7 -> 84.8 | Improved |
| train with 5-6 rephrased prompts instead of a single prompt (ablation) | PARIS3D · Bottle, Knife, Clock, Chair | category mIoU: Bottle 73.8 -> 83.9 | Improved |
| train one model per category instead of a single unified model (ablation) | PARIS3D · Keyboard, Bucket | category mIoU: +34.6% Keyboard, +47.4% Bucket vs PartSLIP | Improved |
| one unified model for all categories vs one per category (PartSLIP) | PartSLIP · PartNet-Ensemble 45 categories | category mIoU: 59.4 -> 29.9 | Hurt |
Later papers that compared with this method
1 later accepted paper reports PARIS3D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Kestrel: 3D Multimodal LLM for Part-Aware Grounded Description (ICCV 2025 · better on 4 of 4 shared results)
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
- MLLM-For3D: Adapting Multimodal Large Language Model for 3D Reasoning Segmentation (NeurIPS 2025 · similarity 0.91)
- PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding (NeurIPS 2025 · similarity 0.89)
- Segment Any 3D Object with Language (ICLR 2025 · similarity 0.89)
- PARTONOMY: Large Multimodal Models with Part-Level Visual Understanding (NeurIPS 2025 · similarity 0.88)
- PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation (CVPR 2024 · similarity 0.88)
- LISA: Reasoning Segmentation via Large Language Model (CVPR 2024 · similarity 0.88)
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.
