Research map / Detection and segmentation / 3D detection and driving
HotBEV: Hardware-oriented Transformer-based Multi-View 3D Detector for BEV Perception
Peiyan Dong, Zhenglun Kong, Xin Meng, Pinrui Yu, Yifan Gong, Geng Yuan, Hao Tang, Yanzhi Wang
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 |
|---|---|---|---|
| insert global attention (GA) after local attention in backbone (ablation) | HotBEV-nano · nuScenes val | NDS: 0.396 -> 0.443 | Improved |
| insert global attention (GA) after local attention in backbone (ablation) | HotBEV-nano · nuScenes val | mAP: 0.328 -> 0.337 | Improved |
| add a second GA block (ablation) | HotBEV-nano · nuScenes val | FPS: 23.7 -> 21.4 | Hurt |
| replace GA with semantic-augmented module (SAM) (ablation) | HotBEV-nano · nuScenes val | NDS: 0.447 -> 0.455 | Improved |
| replace GA with SAM (ablation) | HotBEV-nano · nuScenes val | mAP: 0.340 -> 0.350 | Improved |
| replace GA with SAM (ablation) | HotBEV-nano · nuScenes val | FPS: 21.4 -> 24.5 | Improved |
| finest granularity G=2 in latency model (ablation) | HOB backbone · nuScenes val | FPS | Hurt |
| coarsest granularity G=64 (ablation) | HOB backbone · nuScenes val | mAP | Hurt |
| HOB-tiny vs ResNet50 baselines at 256x704 (BEVDepth) | HotBEV · nuScenes val | NDS: 0.475 -> 0.487 | Improved |
| HOB-tiny vs BEVDet at 256x704 | HotBEV · nuScenes val | mAP: 0.298 -> 0.362 | Improved |
| larger input resolution 512x1408 for HOB-tiny | HotBEV · nuScenes val | NDS: 0.487 -> 0.512 | Improved |
| scale HOB-nano to HOB-base at 256x704 | HotBEV · nuScenes val | NDS: 0.455 -> 0.506 | 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.
- PETRv2 — PETR: Position Embedding Transformation for Multi-View 3D Object Detection (ECCV 2022) · better on 3 of 3 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
- M-BEV: Masked BEV Perception for Robust Autonomous Driving (AAAI 2024 · similarity 0.89)
- BEVSpread: Spread Voxel Pooling for Bird's-Eye-View Representation in Vision-based Roadside 3D Object Detection (CVPR 2024 · similarity 0.87)
- Toward Real-world BEV Perception: Depth Uncertainty Estimation via Gaussian Splatting (CVPR 2025 · similarity 0.87)
- UniMODE: Unified Monocular 3D Object Detection (CVPR 2024 · similarity 0.87)
- BEVDiffuser: Plug-and-Play Diffusion Model for BEV Denoising with Ground-Truth Guidance (CVPR 2025 · similarity 0.87)
- CorrBEV: Multi-View 3D Object Detection by Correlation Learning with Multi-modal Prototypes (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.
