Awesome BEV Perception from Multi-Cameras

ECCV 2020

• LSS: Lift, Splat, Shoot: Encoding Images from Arbitrary Camera Rigs by Implicitly Unprojecting to 3D [paper] [Github]

CoRL 2021

• DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queries [paper] [Github]

CVPR 2021

• CaDDN:Categorical Depth Distribution Network for Monocular 3D Object Detection [paper] [Github]

ICCV 2021

• FIERY: Future Instance Prediction in Bird's-Eye View from Surround Monocular Cameras [paper] [Github]

CVPR 2022

• CVT: Cross-view Transformers for real-time Map-view Semantic Segmentation [paper] [Github]

ICRA 2022

• Translating Images into Maps [paper][Github]

ACMM 2022

• Graph-DETR3D: Rethinking Overlapping Regions for Multi-View 3D Object Detection [paper]

ECCV 2022

• BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers [paper] [Github]
• PETR: Position Embedding Transformation for Multi-View 3D Object Detection [paper][Github]
• ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning [paper][Github]
• SpatialDETR: Robust Scalable Transformer-Based 3D Object Detection from Multi-View Camera Images with Global Cross-Sensor Attention[paper] [Github]

CoRL 2022

• LaRa: Latents and Rays for Multi-Camera Bird's-Eye-View Semantic Segmentation [paper] [Github]

WACV 2023

• BEVSegFormer: Bird’s Eye View Semantic Segmentation From Arbitrary Camera Rigs [paper]

Arxiv 2022

• BEVDet: High-Performance Multi-Camera 3D Object Detection in Bird-Eye-View [paper] [Github]
• BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detection [paper]
• PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images [paper][Github]
• M2BEV: Multi-Camera Joint 3D Detection and Segmentation with Unified Birds-Eye View Representation [paper]
• BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Driving [paper] [Github]
• PolarDETR: Polar Parametrization for Vision-based Surround-View 3D Detection[paper] [Github]
• (AAAI 2023) PolarFormer: Multi-camera 3D Object Detection with Polar Transformers[paper] [Github]
• (ICRA 2023) CrossDTR: Cross-view and Depth-guided Transformers for 3D Object Detection[paper] [Github]
• (AAAI 2023) BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detection [paper][Github]
• A Simple Baseline for BEV Perception Without LiDAR [paper] [Github]
• BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervision [paper]
• AeDet: Azimuth-invariant Multi-view 3D Object Detection [paper] [Github

BEV + Stereo

• (AAAI 2023) BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereo [paper] [Github]
• Time Will Tell: New Outlooks and A Baseline for Temporal Multi-View 3D Object Detection [paper][Github]
• STS: Surround-view Temporal Stereo for Multi-view 3D Detection [paper]

BEV + Distillation

• (ICLR 2023) BEVDistill: Cross-Modal BEV Distillation for Multi-View 3D Object Detection [paper] [Github]
• TiG-BEV: Multi-view BEV 3D Object Detection via Target Inner-Geometry Learning [paper][Github]

Robust BEV

• RoboBEV: Towards Robust Bird's Eye View Detection under Corruptions [paper] [Github]

BEV + Fast

• Fast-BEV: A Fast and Strong Bird’s-Eye View Perception Baseline [paper] [Github]
• MatrixVT: Efficient Multi-Camera to BEV Transformation for 3D Perception [paper][Github]

HD Map Construction

• (ICRA 2022) HDMapNet: An Online HD Map Construction and Evaluation Framework [paper] [Github]
• (ICLR 2023) MapTR: Structured Modeling and Learning for Online Vectorized HD Map Construction [paper] [Github]

Multi-sensor fusion

• FUTR3D: A Unified Sensor Fusion Framework for 3D Detection [paper] [Github]
• (NeurIPS 2022) BEVFusion: A Simple and Robust LiDAR-Camera Fusion Framework [paper] [Github]
• (NeurIPS 2022) Unifying Voxel-based Representation with Transformer for 3D Object Detection [paper] [Github]
• BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation [paper] [Github]
• CMT: Cross Modal Transformer via Coordinates Encoding for 3D Object Dectection [paper] [Github]

Survey

• Vision-Centric BEV Perception: A Survey [paper] [Github]
• Delving into the Devils of Bird's-eye-view Perception: A Review, Evaluation and Recipe [paper][Github]

Occupancy Network

• TPVFormer: An academic alternative to Tesla's Occupancy Network [Github]

Pre-training

• Voxel-MAE: Masked Autoencoders for Self-supervised Pre-training Large-scale Point Clouds [paper][Github]
• BEV-MAE: Bird's Eye View Masked Autoencoders for Outdoor Point Cloud Pre-training[paper][Github]

BEV + Dataset

• aiMotive Dataset: A Multimodal Dataset for Robust Autonomous Driving with Long-Range Perception [paper] [Github]

others

• Focal Sparse Convolutional Networks for 3D Object Detection [paper] [Github]
• Voxel Field Fusion for 3D Object Detection [paper] [Github]
• Scaling up Kernels in 3D CNNs [paper] [Github]

nuScenes detection task Leaderboard