Research map / Detection and segmentation / 3D detection and driving
OPEN: Object-wise Position Embedding for Multi-view 3D Object Detection
Jinghua Hou, Tong Wang, Xiaoqing Ye, Zhe Liu, Shi Gong, Xiao Tan, Errui Ding, Jingdong Wang, Xiang Bai
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 pixel-wise depth supervision (PDE) alone to StreamPETR (ablation) | StreamPETR · nuScenes val | mAP: 50.3 -> 50.5 | No clear effect |
| add object-wise position embedding (OPE) in transformer decoder (ablation) | OPEN · nuScenes val | NDS: 59.7 -> 60.8 | Improved |
| add depth-aware focal loss (DFL) (ablation) | OPEN · nuScenes val | NDS: 60.8 -> 61.3 NDS; 52.4 -> 52.1 mAP | Mixed |
| add object-wise depth encoder (ODE) on top of PDE (ablation) | OPEN · nuScenes val | NDS: 59.4 -> 59.7 | Improved |
| use temporal information in ODE (ablation) | OPEN · nuScenes val | NDS: 61.0 -> 61.3 | Improved |
| add pixel-wise depth encoder as prior for object-wise depth (ablation) | OPEN · nuScenes val | NDS: 59.9 -> 61.3 | Improved |
| object-wise position embedding vs point-aware PE (3DPPE) (ablation) | OPEN · nuScenes val | NDS: 60.0 -> 60.8 | Improved |
| object-wise position embedding vs ray-aware PE on distant objects (ablation) | StreamPETR baseline · nuScenes val, objects beyond 40m | NDS: 36.8 -> 39.1 | Improved |
| predict object-wise depth instead of pixel-wise depth | depth prediction · nuScenes val, objects beyond 40m | L1 error (m): 12.28 -> 3.88 | Improved |
| apply OPEN to StreamPETR with ViT-L backbone at 48 epochs | StreamPETR · nuScenes val (ViT-L) | NDS: 61.3 -> 62.5 | Improved |
| apply OPEN to single-frame PETR | PETR · nuScenes val single frame | NDS: 45.5 -> 50.3 | 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.
- Far3D — Far3D: Expanding the Horizon for Surround-View 3D Object Detection (AAAI 2024) · better on 2 of 2 shared results
- PETR — PETR: Position Embedding Transformation for Multi-View 3D Object Detection (ECCV 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
- Attention-Based Depth Distillation with 3D-Aware Positional Encoding for Monocular 3D Object Detection (AAAI 2023 · similarity 0.91)
- Advancing 3D Object Detection With Depth-Aware Spatial Knowledge Distillation (TPAMI 2025 · similarity 0.90)
- CAPE: Camera View Position Embedding for Multi-View 3D Object Detection (CVPR 2023 · similarity 0.90)
- OpenSight: A Simple Open-Vocabulary Framework for LiDAR-Based Object Detection (ECCV 2024 · similarity 0.90)
- ObjectFusion: Multi-modal 3D Object Detection with Object-Centric Fusion (ICCV 2023 · 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.
