Research map / 3D vision / 3D generation
ALOcc: Adaptive Lifting-Based 3D Semantic Occupancy and Cost Volume-Based Flow Predictions
Dubing Chen, Jin Fang, Wencheng Han, Xinjing Cheng, Junbo Yin, Chengzhong Xu, Fahad Shahbaz Khan, Jianbing Shen
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 adaptive lifting (ablation) | ALOcc-2D-40 · Occ3D with mask | mIoU^m: 43.5 -> 44.5 | Improved |
| add semantic prototype-based occupancy head (vs FBOcc head) (ablation) | ALOcc-2D-40 · Occ3D with mask | mIoU^m: 42.1 -> 44.5 | Improved |
| add semantic prototype head (ablation) | ALOcc-2D-40 · Occ3D with mask | RayIoU: 39.9 -> 42.5 | Improved |
| add adaptive lifting without SP head (ablation) | ALOcc-2D-40 · Occ3D with mask | mIoU^m: 41.2 -> 42.1 | Improved |
| add flow head with simple convolution (ablation) | ALOcc-2D-40 · OpenOcc | RayIoU: 42.4 -> 39.7 | Hurt |
| add bin classification to flow head (ablation) | flow head · OpenOcc | mAVE: 0.597 -> 0.565 | Improved |
| add BEV cost volume to flow head (ablation) | flow head · OpenOcc | Occ Score: 40.7 -> 41.1 | Improved |
| add channel scaling from 40 to 80 (ablation) | ALOcc-Flow-2D · OpenOcc | Occ Score: 41.1 -> 42.1 | Improved |
| add conversion of depth weights to occlusion weights (ablation) | adaptive lifting · Occ3D without mask | mIoU^m: 41.2 -> 41.6 | Improved |
| add depth denoising (ablation) | adaptive lifting · Occ3D without mask | mIoU^m: 41.8 -> 42.1 | Improved |
| set inter-object occlusion m to 3 instead of 0 (ablation) | adaptive lifting · Occ3D | mIoU^m: 41.6 -> 42.1 | Improved |
| use 3D volume encoding instead of 2D height-compressed | ALOcc · Occ3D with mask (ResNet-50) | mIoU^m: 44.8 -> 45.5 | 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.
- COTR — COTR: Correspondence Transformer for Matching Across Images (ICCV 2021) · better on 4 of 4 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
- OctOcc: High-Resolution 3D Occupancy Prediction with Octree (AAAI 2024 · similarity 0.89)
- AdaOcc: Adaptive 3D Occupancy Prediction for Embodied Tasks (NeurIPS 2026 · similarity 0.89)
- VoxelSplat: Dynamic Gaussian Splatting as an Effective Loss for Occupancy and Flow Prediction (CVPR 2025 · similarity 0.89)
- STCOcc: Sparse Spatial-Temporal Cascade Renovation for 3D Occupancy and Scene Flow Prediction (CVPR 2025 · similarity 0.89)
- LowRankOcc: Tensor Decomposition and Low-Rank Recovery for Vision-based 3D Semantic Occupancy Prediction (CVPR 2024 · similarity 0.88)
- 3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation (CVPR 2025 · 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.
