Research map / Detection and segmentation
3D detection and driving: research map
761 papers on 3D detection and driving in Detection and segmentation, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 2 clusters and 7 approaches. The busiest year so far is 2023.
What it is
Self-driving cars must locate cars, people and obstacles in 3D from cameras, LiDAR and radar. This topic covers 3D object detection from each sensor, bird's-eye-view (BEV) representations that combine several cameras, and sensor fusion.
Terms you will see on this page:
- LiDAR
- a laser sensor that measures distances as a 3D point cloud.
- BEV
- bird's-eye view, a top-down map of the scene used by many driving models.
- Monocular 3D detection
- estimating 3D boxes from a single camera image.
Start reading with the most cited papers here:
- nuScenes: A Multimodal Dataset for Autonomous Driving (CVPR 2020 · 6,931 citations)
- Center-Based 3D Object Detection and Tracking (CVPR 2021 · 2,039 citations)
Within Detection and segmentation, its share held steady from 12.6% in 2023–24 to 12.6% in 2025–26 (158 → 152 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore 3D detection and driving in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Camera-radar fusion detection rising: 6% → 12% of the topic (10 → 18 papers)
- LiDAR perception for driving rising: 16% → 21% of the topic (26 → 32 papers)
- Unified LiDAR 3D detection rising: 20% → 21% of the topic (31 → 32 papers)
- 3D objectness and scene segmentation falling: 15% → 10% of the topic (23 → 15 papers)
- Depth-based and label-free 3D detection falling: 14% → 10% of the topic (22 → 15 papers)
Approaches and key papers
3D object detection · 460 papers
Keywords: 3d object, monocular, kitti
Approaches in this cluster:
- Unified LiDAR 3D detection (188 papers)
Detect 3D objects from point clouds with grouping, graphs, and domain adaptation. - Depth-based and label-free 3D detection (91 papers)
Detect 3D objects using pseudo-LiDAR depth, stereo, and label-free mobile object discovery. - 3D objectness and scene segmentation (79 papers)
Learn open-world 3D objectness and instance segmentation with self-supervised and contrastive methods. - Monocular 3D object detection (102 papers)
Improve single-image 3D detection with auxiliary contexts, distillation, and depth-aware losses.
Most cited and most cited since 2024:
- Center-Based 3D Object Detection and Tracking (CVPR 2021 · 2,039 citations)
- 3D Bounding Box Estimation Using Deep Learning and Geometry (CVPR 2017 · 1,208 citations)
- SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection (CVPR 2024 · 78 citations)
- Far3D: Expanding the Horizon for Surround-View 3D Object Detection (AAAI 2024 · 73 citations)
BEV perception and sensor fusion · 301 papers
Keywords: bev, autonomous, radar
Approaches in this cluster:
- LiDAR perception for driving (115 papers)
Build efficient LiDAR detectors, anomaly datasets, and generative models for driving perception. - Multi-camera BEV detection (129 papers)
Detect 3D objects from multiple cameras with BEV representations and cross-view attention. - Camera-radar fusion detection (57 papers)
Fuse radar with camera or LiDAR for robust 3D vehicle detection.
Most cited and most cited since 2024:
- nuScenes: A Multimodal Dataset for Autonomous Driving (CVPR 2020 · 6,931 citations)
- TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection With Transformers (CVPR 2022 · 939 citations)
- RCBEVDet: Radar-camera Fusion in Bird's Eye View for 3D Object Detection (CVPR 2024 · 133 citations)
- IS-Fusion: Instance-Scene Collaborative Fusion for Multimodal 3D Object Detection (CVPR 2024 · 118 citations)
Related topics in Detection and segmentation
- Tracking (575)
- Open-set and few-shot detection (2,864)
- CNN features (2,941)
- Weakly supervised detection and segmentation (791)
