Research map / 3D vision / Gaussian splatting
S2GO: Streaming Sparse Gaussian Occupancy
Jinhyung Park, Chensheng Peng, yihan hu, Wenzhao Zheng, Kris Kitani, Wei Zhan
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 denoising loss in pretraining (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 20.55 -> 21.60 | Improved |
| add RGB loss in pretraining (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 20.25 -> 20.55 | Improved |
| initialize queries from noised LiDAR vs learnable (ablation) | S2GO pretraining · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 12.42 -> 20.55 | Improved |
| initialize queries from exact LiDAR instead of noised LiDAR (ablation) | S2GO pretraining · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 20.55 -> 13.62 | Hurt |
| add pretraining (vs 24 epochs direct) (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 15.83 -> 21.60 | Improved |
| add temporal query propagation (top-k opacity) (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 17.92 -> 19.94 | Improved |
| use delta-distance separation in top-k propagation (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 19.94 -> 20.51 | Improved |
| add velocity modeling in pretraining (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 20.15 -> 20.55 | Improved |
| add velocity modeling in occupancy estimation stage (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 20.07 -> 20.15 | Improved |
| include opacity in occupancy probability (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 16.97 -> 20.13 | Improved |
| use efficient Gaussian-to-voxel splatting (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | Train GPU hours: 129h -> 28h | Improved |
| increase queries and Gaussians 900x10 to 1800x20 (ablation) | S2GO · nuScenes-SurroundOcc val (ablation, 12+12 epochs) | mIoU: 21.60 -> 21.84 | Improved |
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.
- GaussianWorld — GaussianWorld: Gaussian World Model for Streaming 3D Occupancy Prediction (CVPR 2025) · better on 3 of 3 shared results
- GaussianFormer-2 — GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Prediction (CVPR 2025) · better on 2 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
- ODG: Occupancy Prediction Using Dual Gaussians (NeurIPS 2025 · similarity 0.91)
- GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Prediction (CVPR 2025 · similarity 0.89)
- GaussianWorld: Gaussian World Model for Streaming 3D Occupancy Prediction (CVPR 2025 · similarity 0.89)
- VoxelSplat: Dynamic Gaussian Splatting as an Effective Loss for Occupancy and Flow Prediction (CVPR 2025 · similarity 0.88)
- GaussTR: Foundation Model-Aligned Gaussian Transformer for Self-Supervised 3D Spatial Understanding (CVPR 2025 · similarity 0.88)
- STCOcc: Sparse Spatial-Temporal Cascade Renovation for 3D Occupancy and Scene Flow Prediction (CVPR 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.
