Research map / 3D vision / Gaussian splatting
GaussianFlowOcc: Sparse and Weakly Supervised Occupancy Estimation using Gaussian Splatting and Temporal Flow
Simon Boeder, Fabian Gigengack, Benjamin Risse
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 induced self-attention and induced temporal attention (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 13.81 -> 17.08 | Improved |
| add induced self-attention alone (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 13.81 -> 14.60 | Improved |
| add induced temporal attention alone (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 13.81 -> 14.47 | Improved |
| add the Temporal Module (3D flow per Gaussian) (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 14.18 -> 17.08 | Improved |
| add the Temporal Module (3D flow per Gaussian) (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | RayIoU: 14.46 -> 16.47 | Improved |
| learn opacity instead of fixing it to 1 (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 9.48 -> 13.12 | Improved |
| also learn scale (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 13.12 -> 14.98 | Improved |
| also learn rotation (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 14.98 -> 17.08 | Improved |
| increase temporal horizon T from 0 to 6 (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 12.67 -> 17.08 | Improved |
| increase temporal horizon T from 6 to 8 (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 17.08 -> 14.82 | Hurt |
| add Metric3D pseudo depth supervision (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | mIoU: 14.07 -> 17.08 | Improved |
| add Metric3D pseudo depth supervision (ablation) | GaussianFlowOcc · Occ3D-nuScenes val | IoU: 34.12 -> 46.91 | Improved |
Later papers that compared with this method
1 later accepted paper reports GaussianFlowOcc as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- ShelfOcc: Native 3D Supervision beyond LiDAR for Vision-Based Occupancy Estimation (CVPR 2026 · better on 3 of 3 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)
- VoxelSplat: Dynamic Gaussian Splatting as an Effective Loss for Occupancy and Flow Prediction (CVPR 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)
- GaussTR: Foundation Model-Aligned Gaussian Transformer for Self-Supervised 3D Spatial Understanding (CVPR 2025 · similarity 0.88)
- SparseOcc: Rethinking Sparse Latent Representation for Vision-Based Semantic Occupancy Prediction (CVPR 2024 · 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.
