Camera-Only Bird's Eye View Perception: A Neural Approach to LiDAR-Free Environmental Mapping for Autonomous Vehicles

2025/05/09 by Anupkumar Bochare, Bochare, Anupkumar
Computer Science · Engineering · #Advanced Neural Network Applications #Autonomous Vehicle Technology and Safety #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Robotics and Sensor-Based Localization

paper · pdf · doi:10.48550/arxiv.2505.06113

openalex publication_date 2025/05/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Autonomous vehicle perception systems have traditionally relied on costly LiDAR sensors to generate precise environmental representations. In this paper, we propose a camera-only perception framework that produces Bird's Eye View (BEV) maps by extending the Lift-Splat-Shoot architecture. Our method combines YOLOv11-based object detection with DepthAnythingV2 monocular depth estimation across multi-camera inputs to achieve comprehensive 360-degree scene understanding. We evaluate our approach on the OpenLane-V2 and NuScenes datasets, achieving up to 85% road segmentation accuracy and 85-90% vehicle detection rates when compared against LiDAR ground truth, with average positional errors limited to 1.2 meters. These results highlight the potential of deep learning to extract rich spatial information using only camera inputs, enabling cost-efficient autonomous navigation without sacrificing accuracy.

Citations

• BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers
• Vision Transformers for Dense Prediction
• Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D
• nuScenes: A multimodal dataset for autonomous driving
• PointPillars: Fast Encoders for Object Detection from Point Clouds
• MVSNet: Depth Inference for Unstructured Multi-view Stereo
• Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
• Depth Map Prediction from a Single Image using a Multi-Scale Deep Network

Related

• Lift-Attend-Splat: Bird's-eye-view camera-lidar fusion using transformers
• Multi-camera Bird's Eye View Perception for Autonomous Driving
• Vision-Driven 2D Supervised Fine-Tuning Framework for Bird's Eye View Perception
• Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D
• BirdNet: a 3D Object Detection Framework from LiDAR information