Research map / 3D vision / 3D generation
SA-Occ: Satellite-Assisted 3D Occupancy Prediction in Real World
Chen Chen, Zhirui Wang, Taowei Sheng, Yi Jiang, Yundu Li, Peirui Cheng, Luning Zhang, Kaiqiang Chen, Yanfeng Hu, Xue Yang, Xian Sun
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 satellite BEV branch (ablation) | FlashOcc baseline · Occ3D-nuScenes | mIoU: 32.08 -> 36.89 (+ 4.81) | Improved |
| add Uniform Sampling Alignment to street branch (ablation) | SA-Occ with satellite branch · Occ3D-nuScenes | mIoU: 36.89 -> 38.55 | Improved |
| add Dynamic-Decoupling Fusion (ablation) | SA-Occ with satellite branch and Uni-SA · Occ3D-nuScenes | mIoU: 38.55 -> 39.05 (+ 0.50) | Improved |
| add Dynamic-Decoupling Fusion without Uni-SA (ablation) | SA-Occ with satellite branch · Occ3D-nuScenes | mIoU: 36.89 -> 37.16 | Improved |
| add DDF (ablation) | SA-Occ with satellite branch and Uni-SA · Occ3D-nuScenes | S mIoU: 54.93 -> 54.55 | Hurt |
| use DDF instead of add fusion (ablation) | SA-Occ · Occ3D-nuScenes | mIoU: 38.57 -> 39.05 | Improved |
| use DDF instead of MaskSegAtt attention fusion (ablation) | SA-Occ · Occ3D-nuScenes | mIoU: 23.23 -> 39.05 | Improved |
| add alignment operation to DDF (ablation) | SA-Occ · Occ3D-nuScenes | mIoU: 39.05 -> 38.78 | Hurt |
| add 3D-Proj Guidance to satellite branch (ablation) | SA-Occ satellite branch · Occ3D-nuScenes | mIoU: 36.6 -> 36.82 | Improved |
| add soft gated convolution (ablation) | SA-Occ satellite branch · Occ3D-nuScenes | mIoU: 36.82 -> 36.89 | Improved |
| add Dynamic-Decoupling strategy (ablation) | SA-Occ DDF · Occ3D-nuScenes | mIoU: 38.55 -> 38.88 | Improved |
| add dynamic-encoding spatial attention instead of plain spatial attention (ablation) | SA-Occ DDF · Occ3D-nuScenes | mIoU: 38.95 -> 39.05 | 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.
- OccFormer — OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Prediction (ICCV 2023) · better on 1 of 1 shared result
- MonoScene — MonoScene: Monocular 3D Semantic Scene Completion (CVPR 2022) · 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
- Spatiotemporal Decoupling for Efficient Vision-Based Occupancy Forecasting (CVPR 2025 · similarity 0.91)
- Cam4DOcc: Benchmark for Camera-Only 4D Occupancy Forecasting in Autonomous Driving Applications (CVPR 2024 · similarity 0.89)
- SDGOCC: Semantic and Depth-Guided Bird's-Eye View Transformation for 3D Multimodal Occupancy Prediction (CVPR 2025 · similarity 0.89)
- LOMA: Language-assisted Semantic Occupancy Network via Triplane Mamba (AAAI 2025 · similarity 0.88)
- OccProphet: Pushing the Efficiency Frontier of Camera-Only 4D Occupancy Forecasting with an Observer-Forecaster-Refiner Framework (ICLR 2025 · similarity 0.87)
- COTR: Compact Occupancy TRansformer for Vision-based 3D Occupancy Prediction (CVPR 2024 · 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.
