Research map / 3D vision / Gaussian splatting
SkySplat-OV: Generalizable Language Gaussian Splatting for Open-Vocabulary Scene Understanding from Sparse Satellite Views
Xuejun Huang, Yi Wan, Lei Yu, Xinyi Liu, Zhi Zheng, Bin Zhang, Yingying Pei, Yi Liu, Changjun Zhu, Yi Liu, Xiangyuan Cai, Hongwei Hu and 2 more
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 GRS-Fusion module (ablation) | SkySplat-OV · DFC19 | mIoU: 46.38 -> 48.12 | Improved |
| add relative height supervision (ablation) | SkySplat-OV · DFC19 | mIoU: 43.12 -> 48.12 | Improved |
| add relative height supervision (ablation) | SkySplat-OV · DFC19 | PSNR: 17.64 -> 17.63 | No clear effect |
| distill SAM3 features into 3DGS instead of SAM3 on rendered views (ablation) | SkySplat-OV · DFC19 | mIoU: 31.97 -> 48.12 | Improved |
| add feature distillation (vs SkySplat + SAM3) (ablation) | SkySplat · DFC19 | mIoU: 22.44 -> 48.12 | Improved |
| use RPC camera model instead of pinhole (ablation) | SkySplat-OV · DFC19 | mIoU: 38.45 -> 48.12 | Improved |
| SkySplat-OV (pinhole) vs LSM/Uni3R (ablation) | SkySplat-OV · DFC19 | mIoU: 22.20 -> 38.45 | Improved |
| feed-forward SkySplat-OV vs per-scene optimization LangSplatV2 | SkySplat-OV · 8 AOIs | time: 2.92h -> 9.82s | Improved |
| feed-forward SkySplat-OV vs Feature-3DGS | SkySplat-OV · 8 AOIs | mean mIoU: 22.71 -> 42.15 | Improved |
| SkySplat-OV vs HiSplat height reconstruction on MVS3D | SkySplat-OV · DFC19 -> MVS3D | MAE: 18.01 -> 4.28 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- Uni3R — Uni3R: Unified 3D Reconstruction and Semantic Understanding via Generalizable Gaussian Splatting from Unposed Multi-View Images (CVPR 2026) · better on 4 of 4 shared results
- LangSplatV2 — LangSplatV2: High-dimensional 3D Language Gaussian Splatting with 450+ FPS (NeurIPS 2025) · better on 1 of 2 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
- SkySense-O: Towards Open-World Remote Sensing Interpretation with Vision-Centric Visual-Language Modeling (CVPR 2025 · similarity 0.88)
- SceneSplat++: A Large Dataset and Comprehensive Benchmark for Language Gaussian Splatting (NeurIPS 2025 · similarity 0.88)
- Tackling View-Dependent Semantics in 3D Language Gaussian Splatting (ICML 2025 · similarity 0.88)
- SkyMoE: A Vision-Language Foundation Model for Enhancing Geospatial Interpretation with Mixture of Experts (AAAI 2026 · similarity 0.88)
- Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understanding (CVPR 2025 · similarity 0.87)
- SOVGaussian: Sparse-View 3D Gaussian Splatting for Open-Vocabulary Scene Understanding (AAAI 2025 · similarity 0.87)
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.
