Research map / Detection and segmentation / 3D detection and driving
ObjectFusion: Multi-modal 3D Object Detection with Object-Centric Fusion
Qi Cai, Yingwei Pan, Ting Yao, Chong-Wah Ngo, Tao Mei
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 |
|---|---|---|---|
| use all three object-centric features (BEV, voxel, image) instead of one (ablation) | ObjectFusion · nuScenes val | mAP (%): 69.3 -> 69.8 | Improved |
| add modality-specific context encoder (MSCE) (ablation) | ObjectFusion · nuScenes val | mAP (%): 69.3 -> 69.8 | Improved |
| add MSCE (NDS) (ablation) | ObjectFusion · nuScenes val | NDS (%): 71.5 -> 72.3 | Improved |
| replace BEV-based fusion with object-centric fusion | BEVFusion · nuScenes val | mAP (%): 68.5 -> 69.8 | Improved |
| object-centric fusion instead of BEV fusion in night scenes | BEVFusion · nuScenes val, Night | mAP (%): 42.8 -> 46.0 | Improved |
| add translation offsets to camera-LiDAR calibration (with RoI enlarging) (ablation) | ObjectFusion · nuScenes val | mAP (%): 69.8 -> 69.3 at 1.0m offset | No clear effect |
| object-centric fusion on Waymo versus TransFusion | TransFusion · Waymo val | L2 mAPH (%): 65.5 -> 66.3 | Improved |
| extract and fuse object-centric features | BEVFusion · nuScenes val | inference time (ms): 257 -> 274 | Hurt |
| object-centric fusion instead of BEV fusion for tracking | BEVFusion · nuScenes tracking val | AMOTA (%): 72.8 -> 74.2 | Improved |
Later papers that compared with this method
1 later accepted paper reports ObjectFusion as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- GraphBEV: Towards Robust BEV Feature Alignment for Multi-Modal 3D Object Detection (ECCV 2024 · better on 2 of 2 shared results)
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.
- TransFusion — TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection With Transformers (CVPR 2022) · better on 2 of 2 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
- DeepFusion: Lidar-Camera Deep Fusion for Multi-Modal 3D Object Detection (CVPR 2022 · similarity 0.93)
- SparseFusion: Fusing Multi-Modal Sparse Representations for Multi-Sensor 3D Object Detection (ICCV 2023 · similarity 0.92)
- BEVDilation: LiDAR-Centric Multi-Modal Fusion for 3D Object Detection (AAAI 2026 · similarity 0.91)
- GAFusion: Adaptive Fusing LiDAR and Camera with Multiple Guidance for 3D Object Detection (CVPR 2024 · similarity 0.91)
- SimDistill: Simulated Multi-Modal Distillation for BEV 3D Object Detection (AAAI 2024 · similarity 0.90)
- GeoBEV: Learning Geometric BEV Representation for Multi-view 3D Object Detection (AAAI 2025 · similarity 0.89)
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.
