Research map / 3D vision / Gaussian splatting
ReLaGS: Relational Language Gaussian Splatting
Yaxu Xie, Abdalla Arafa, Alireza Javanmardi, Christen Millerdurai, Jia Cheng Hu, Shaoxiang Wang, Alain Pagani, Didier Stricker
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 Maximum Weight Pruning (ablation) | ReLaGS base · LERF-OVS | mean mIoU: 55.88 -> 62.04 | Improved |
| add Robust Outlier-Aware Feature Aggregation (ablation) | ReLaGS Base+MWP · LERF-OVS | mean mIoU: 62.04 -> 64.36 | Improved |
| add Maximum Weight Pruning (ablation) | ReLaGS base · 3DSSG | Object Recall@5: 0.57 -> 0.65 | Improved |
| add Robust Outlier-Aware Feature Aggregation (ablation) | ReLaGS Base+MWP · 3DSSG | Object Recall@5: 0.65 -> 0.68 | Improved |
| add Robust Outlier-Aware Feature Aggregation (ablation) | ReLaGS Base+MWP · 3DSSG | Predicate Recall@3: 0.77 -> 0.79 | Improved |
| apply light pruning threshold 5e-4 (ablation) | ReLaGS · LERF-OVS | mIoU: 57.84 -> 64.36 | Improved |
| raise pruning threshold to 5e-1 (ablation) | ReLaGS · LERF-OVS | mIoU: 64.36 -> 39.18 | Hurt |
| raise tau_lang from 3 to 4 (ablation) | ReLaGS · LERF-OVS | mean mIoU: 64.36 -> 63.07 | Hurt |
| lower tau_lang from 3 to 1 (ablation) | ReLaGS · LERF-OVS | mean mIoU: 64.36 -> 60.78 | Hurt |
| add multi-hierarchy search (ablation) | ReLaGS w/o MH · ScanNet++ | mIoU: 0.54 -> 0.56 | Improved |
| use two transformer layers instead of one in GNN (ablation) | ReLaGS GNN · 3DSSG | Predicate Recall@3: 0.79 -> 0.77 | No clear effect |
| initialize edge features from node pair instead of zeros (ablation) | ReLaGS GNN · 3DSSG | Predicate Recall@3: 0.75 -> 0.79 | Mixed |
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.
- LangSplat — LangSplat: 3D Language Gaussian Splatting (CVPR 2024) · better on 1 of 1 shared result
- LERF — LERF: Language Embedded Radiance Fields (ICCV 2023) · 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 0.91)
- Chain of Semantics Programming in 3D Gaussian Splatting Representation for 3D Vision Grounding (CVPR 2025 · similarity 0.91)
- Language Embedded 3D Gaussians for Open-Vocabulary Scene Understanding (CVPR 2024 · similarity 0.90)
- Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration (CVPR 2025 · similarity 0.90)
- SOVGaussian: Sparse-View 3D Gaussian Splatting for Open-Vocabulary Scene Understanding (AAAI 2025 · similarity 0.89)
- ReferSplat: Referring Segmentation in 3D Gaussian Splatting (ICML 2025 · similarity 0.89)
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.
