Research map / 3D vision / Gaussian splatting
Online Language Splatting
Saimouli Katragadda, Cho-Ying Wu, Yuliang Guo, Xinyu Huang, Guoquan Huang, Liu Ren
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 super-resolution decoder (SRD, Omnidata-trained) to CLIP embedding (ablation) | Online Language Splatting · Replica (text query localization) | mIoU: 0.400 -> 0.485 | Improved |
| add super-resolution decoder (SRD, Omnidata-trained) to CLIP embedding (ablation) | Online Language Splatting · Replica (text query localization) | Loc accuracy: 0.754 -> 0.802 | Improved |
| add online-learned autoencoder (OLAE) with Omnidata SRD (ablation) | Online Language Splatting · Replica (text query localization) | Loc accuracy: 0.802 -> 0.826 | Improved |
| add online-learned autoencoder (OLAE) with COCO SRD (ablation) | Online Language Splatting · Replica (text query localization) | Loc accuracy: 0.782 -> 0.759 | Hurt |
| add online-learned autoencoder (OLAE) with COCO SRD (ablation) | Online Language Splatting · Replica (text query localization) | mIoU: 0.475 -> 0.479 | No clear effect |
| use color-language disentangled optimization instead of joint multi-channel (ablation) | Online Language Splatting · Replica Room-0 (disentanglement ablation) | mIoU: 0.323 -> 0.402 | Improved |
| use color-language disentangled optimization instead of joint multi-channel (ablation) | Online Language Splatting · Replica Room-0 (disentanglement ablation) | PSNR: 31.23 -> 35.89 | Improved |
| use color-language disentangled optimization instead of joint multi-channel (ablation) | Online Language Splatting · Replica Room-0 (disentanglement ablation) | ATE (cm): 0.796 -> 0.325 | Improved |
| use color-language disentangled optimization instead of joint multi-channel (ablation) | Online Language Splatting · Replica Room-0 (disentanglement ablation) | 2D Loc accuracy: 0.633 -> 0.622 | No clear effect |
| use color-language disentangled optimization instead of joint multi-channel (ablation) | Online Language Splatting · Replica Room-0 (disentanglement ablation) | 3D Chamfer distance: 1.940 -> 0.974 | Improved |
| add generalized encoder for cross-view consistency (ablation) | Online Language Splatting · Replica | cosine similarity across views: drops to 0.5 without it | Improved |
| replace offline LangSplat with online language splatting | Online Language Splatting vs LangSplat · TUM RGB-D Scene1 | Loc accuracy: 0.850 -> 0.917 | Improved |
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.
- LangSplat — LangSplat: 3D Language Gaussian Splatting (CVPR 2024) · better on 6 of 8 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
- LangSplat: 3D Language Gaussian Splatting (CVPR 2024 · similarity 0.92)
- Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration (CVPR 2025 · similarity 0.90)
- LangSplatV2: High-dimensional 3D Language Gaussian Splatting with 450+ FPS (NeurIPS 2025 · similarity 0.90)
- SceneSplat++: A Large Dataset and Comprehensive Benchmark for Language Gaussian Splatting (NeurIPS 2025 · similarity 0.90)
- Language Embedded 3D Gaussians for Open-Vocabulary Scene Understanding (CVPR 2024 · similarity 0.88)
- Segment then Splat: Unified 3D Open-Vocabulary Segmentation via Gaussian Splatting (NeurIPS 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.
