Research map / 3D vision / Gaussian splatting
Uni3R: Unified 3D Reconstruction and Semantic Understanding via Generalizable Gaussian Splatting from Unposed Multi-View Images
Xiangyu Sun, Haoyi Jiang, Liu Liu, Seungtae Nam, Gyeongjin Kang, Xinjie Wang, Wei Sui, Zhizhong Su, Wenyu Liu, Xinggang Wang, Eunbyung Park
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 |
|---|---|---|---|
| add semantic loss (LSeg distillation) (ablation) | Uni3R · ScanNet | mIoU: 0.0183 -> 0.5484 | Improved |
| add semantic loss (LSeg distillation) (ablation) | Uni3R · ScanNet | PSNR: 25.38 -> 25.53 | No clear effect |
| add rendering loss (ablation) | Uni3R · ScanNet | convergence | Improved |
| add point-map geometry loss (frozen VGGT) (ablation) | Uni3R · ScanNet | depth rel error: 5.6 -> 3.9 | Improved |
| add point-map geometry loss (frozen VGGT) (ablation) | Uni3R · ScanNet | depth tau: 48.2 -> 61.2 | Improved |
| add point-map geometry loss (frozen VGGT) (ablation) | Uni3R · ScanNet | PSNR/mIoU: 25.53 -> 25.53 PSNR; 0.5541 -> 0.5484 mIoU | No clear effect |
| add scale-invariant constraint (ablation) | Uni3R · ScanNet | PSNR: 24.95 -> 25.53 | Improved |
| add intrinsic embedding (ablation) | Uni3R · ScanNet | depth tau: 42.5 -> 61.2 | Improved |
| fine-tune transformer layers instead of freezing all (ablation) | Uni3R · ScanNet | PSNR: 5.49 -> 25.53 | Improved |
| use Uni3R vs LSM for 2-view ScanNet | Uni3R · ScanNet 2 views | depth rel error / tau: 3.38 -> 3.87 rel; 67.77 -> 61.37 tau | Hurt |
| use Uni3R vs LSM for 2-view ScanNet | Uni3R · ScanNet 2 views | PSNR and target mIoU: 24.39 -> 25.53; 0.5078 -> 0.5584 | Improved |
| use Uni3R vs NoPoSplat on RE10K 2 views | Uni3R · RE10K 2 views | PSNR: 25.036 -> 25.074 | No clear effect |
Later papers that compared with this method
4 later accepted papers report Uni3R as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- DINOv3D: 2D-3D Joint Optimization for Unified Spatial Understanding (ECCV 2026 · better on 6 of 6 shared results)
- EPS3D: End-to-End Feed-Forward 3D Panoptic Segmentation (ICML 2026 · better on 5 of 5 shared results)
- FLEG: Feed-Forward Language Embedded Gaussian Splatting from Any Views via Compact Semantic Representation (ECCV 2026 · better on 5 of 5 shared results)
- SkySplat-OV: Generalizable Language Gaussian Splatting for Open-Vocabulary Scene Understanding from Sparse Satellite Views (ECCV 2026 · better on 4 of 4 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- PixelSplat — pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024) · better on 2 of 2 shared results
- MVSplat — MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images (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
- Tackling View-Dependent Semantics in 3D Language Gaussian Splatting (ICML 2025 · similarity 0.92)
- SIU3R: Simultaneous Scene Understanding and 3D Reconstruction Beyond Feature Alignment (NeurIPS 2025 · similarity 0.91)
- Large Spatial Model: End-to-end Unposed Images to Semantic 3D (NeurIPS 2024 · similarity 0.90)
- 3D Vision-Language Gaussian Splatting (ICLR 2025 · similarity 0.90)
- Know Your Neighbors: Improving Single-View Reconstruction via Spatial Vision-Language Reasoning (CVPR 2024 · similarity 0.90)
- A Unified Framework for 3D Scene Understanding (NeurIPS 2024 · similarity 0.90)
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.
