Research map / Point clouds / Segmentation and detection
InfoFocus: 3D Object Detection for Autonomous Driving with Dynamic Information Modeling
Jun Wang, Shiyi Lan, Mingfei Gao, Larry S. Davis
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 PoI Pooling (ablation) | PointPillars · nuScenes val | mAP: 29.5 -> 32.5 | Improved |
| add Visibility Attention on top of PoI Pooling (ablation) | InfoFocus · nuScenes val | mAP: 32.5 -> 34.8 | Improved |
| add Adaptive Point-wise Attention on top of PoI Pooling (ablation) | InfoFocus · nuScenes val | mAP: 32.5 -> 34.8 | Improved |
| add Adaptive Point-wise Attention on top of PoIPool + Vis. Att. (ablation) | InfoFocus · nuScenes val | mAP: 34.8 -> 36.4 | Improved |
| use four key-points per edge instead of two (ablation) | InfoFocus · nuScenes val | mAP: degrades slightly by 0.8% mAP | Hurt |
| add rotated RoIAlign 4x4 to PointPillars | PointPillars · nuScenes val | mAP: 29.5 -> 33.9 | Improved |
| InfoFocus instead of rotated RoIAlign 4x4 | InfoFocus · nuScenes val | mAP: 33.9 -> 36.4 | Improved |
| larger rotated RoIAlign pooling size 8x4 vs 4x4 | rotated RoIAlign · nuScenes val | mAP: 33.9 -> 34.3 | Improved |
| double the number of training epochs | InfoFocus · nuScenes test | mAP: 37.5 -> 39.5 | Improved |
| add InfoFocus second stage to PointPillars | PointPillars · nuScenes val | mAP: 29.5 -> 36.4 | Improved |
| add InfoFocus second stage | PointPillars · nuScenes inference time | inference time (ms): 26.9 -> 32.9 | Hurt |
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.
- PointPillars — PointPillars: Fast Encoders for Object Detection From Point Clouds (CVPR 2019) · better on 12 of 12 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
- 3D Object Detection for Autonomous Driving: A Comprehensive Survey (IJCV 2023 · similarity 0.89)
- 3D Object Detection From Images for Autonomous Driving: A Survey (TPAMI 2023 · similarity 0.88)
- DriveGEN: Generalized and Robust 3D Detection in Driving via Controllable Text-to-Image Diffusion Generation (CVPR 2025 · similarity 0.87)
- Ev-3DOD: Pushing the Temporal Boundaries of 3D Object Detection with Event Cameras (CVPR 2025 · similarity 0.87)
- CorrBEV: Multi-View 3D Object Detection by Correlation Learning with Multi-modal Prototypes (CVPR 2025 · similarity 0.87)
- OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection (NeurIPS 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.
