Research map / Detection and segmentation / 3D detection and driving
RIOcc: Efficient Cross-Modal Fusion Transformer with Collaborative Feature Refinement for 3D Semantic Occupancy Prediction
Baojie Fan, Xiaotian Li, Yuhan Zhou, Yuyu Jiang, Jiandong Tian, Huijie Fan
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 Wavelet Encoder (with Semantic Encoder already present) (ablation) | RIOcc · Occ3D-nuScenes | mIoU: 53.37 -> 54.21 | Improved |
| add Semantic Encoder (with Wavelet Encoder already present) (ablation) | RIOcc · Occ3D-nuScenes | mIoU: 53.32 -> 54.21 | Improved |
| add both Wavelet and Semantic Encoders to the baseline (ablation) | RIOcc · Occ3D-nuScenes | mIoU: 51.96 -> 54.21 | Improved |
| add channel-wise attention only to DBP (ablation) | RIOcc · Occ3D-nuScenes | mIoU: 53.07 -> 53.70 | Improved |
| add grid-wise attention only to DBP (ablation) | RIOcc · Occ3D-nuScenes | mIoU: 53.07 -> 53.78 | Improved |
| combine channel-wise and grid-wise branches (full DBP) (ablation) | RIOcc · Occ3D-nuScenes | mIoU: 53.07 -> 54.21 | Improved |
| increase downsampling layers from 0 to 2 (ablation) | RIOcc (no wavelet/semantic encoders) · Occ3D-nuScenes | mIoU: 48.21 -> 52.03 | Improved |
| increase downsampling layers from 2 to 3 (ablation) | RIOcc (no wavelet/semantic encoders) · Occ3D-nuScenes | mIoU: 52.03 -> 52.15 | No clear effect |
| aggregate local regions in DDA: 1x1 to 5x5 (ablation) | RIOcc DDA module · Occ3D-nuScenes | mIoU: 47.43 -> 48.21 | Improved |
| enlarge aggregation region from 5x5 to 7x7 (ablation) | RIOcc DDA module · Occ3D-nuScenes | mIoU: 48.21 -> 48.16 | Hurt |
| replace addition fusion with DDA (ablation) | RIOcc BEV fusion · Occ3D-nuScenes | mIoU: 46.58 -> 48.21 | Improved |
| replace concatenation fusion with DDA (ablation) | RIOcc BEV fusion · Occ3D-nuScenes | mIoU: 46.69 -> 48.21 | 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.
- PanoOcc — PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentation (CVPR 2024) · 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
- SDGOCC: Semantic and Depth-Guided Bird's-Eye View Transformation for 3D Multimodal Occupancy Prediction (CVPR 2025 · similarity 0.92)
- CSV-Occ: Fusing Multi-frame Alignment for Occupancy Prediction with Temporal Cross State Space Model and Central Voting Mechanism (ICML 2025 · similarity 0.88)
- RadOcc: Learning Cross-Modality Occupancy Knowledge through Rendering Assisted Distillation (AAAI 2024 · similarity 0.87)
- Achieving Speed-Accuracy Balance in Vision-based 3D Occupancy Prediction via Geometric-Semantic Disentanglement (AAAI 2025 · similarity 0.87)
- COTR: Compact Occupancy TRansformer for Vision-based 3D Occupancy Prediction (CVPR 2024 · similarity 0.87)
- Rethinking Temporal Fusion with a Unified Gradient Descent View for 3D Semantic Occupancy 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.
