Research map / 3D vision / 3D generation
IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstruction
Hao Li, Zhengyu Zou, Fangfu Liu, Xuanyang Zhang, Fangzhou Hong, Yukang Cao, Yushi LAN, Manyuan Zhang, Gang YU, Dingwen Zhang, Ziwei Liu
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 |
|---|---|---|---|
| joint training of geometry and instance heads (IGGT) vs VGGT geometry only | IGGT · ScanNet++ depth | Abs. Rel: 2.75 -> 2.61 | Improved |
| joint training of geometry and instance heads (IGGT) vs VGGT geometry only | IGGT · ScanNet depth | Abs. Rel: 1.84 -> 1.90 | No clear effect |
| use instance-grounded clustering masks with OpenSeg instead of LSM (language-aligned features) | IGGT · ScanNet++ open-vocab segmentation | 2D mIoU: 17.88 -> 31.31 | Improved |
| instance-grounded 3D-consistent features for tracking vs SAM2* | IGGT · ScanNet spatial tracking | T-SR: 71.25 -> 98.66 | Improved |
| add cross-modal fusion block to instance head (ablation) | IGGT instance head · training and PCA visualization | convergence and edge sharpness | Improved |
| swap VLM plugged into instance masks (LSeg/OpenSeg vs CLIP) (ablation) | IGGT · open-vocabulary segmentation | qualitative accuracy | Mixed |
Later papers that compared with this method
3 later accepted papers report IGGT as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer (NeurIPS 2026 · better on 4 of 4 shared results)
- SegVGGT: Joint 3D Reconstruction and Instance Segmentation from Multi-View Images (ECCV 2026 · better on 4 of 4 shared results)
- Pano3D: Unified 3D Reconstruction and Panoptic Segmentation (ECCV 2026 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 3 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
- IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer (NeurIPS 2026 · similarity 0.92)
- VGGT: Visual Geometry Grounded Transformer (CVPR 2025 · similarity 0.89)
- Inst3D-LMM: Instance-Aware 3D Scene Understanding with Multi-modal Instruction Tuning (CVPR 2025 · similarity 0.89)
- Geometry-Guided Semantic Reconstruction for 3D Instance Segmentation (NeurIPS 2026 · similarity 0.88)
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 2024 · similarity 0.88)
- Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understanding (CVPR 2025 · 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.
