A Resource Efficient Fusion Network for Object Detection in Bird's-Eye View using Camera and Raw Radar Data
This work addresses the challenge of robust perception in adverse weather conditions for autonomous vehicles, representing an incremental improvement in sensor fusion techniques.
The paper tackles the problem of object detection in autonomous driving by fusing camera images and raw radar data in Bird's-Eye View, achieving improved accuracy with reduced computational complexity compared to existing methods on the RADIal dataset.
Cameras can be used to perceive the environment around the vehicle, while affordable radar sensors are popular in autonomous driving systems as they can withstand adverse weather conditions unlike cameras. However, radar point clouds are sparser with low azimuth and elevation resolution that lack semantic and structural information of the scenes, resulting in generally lower radar detection performance. In this work, we directly use the raw range-Doppler (RD) spectrum of radar data, thus avoiding radar signal processing. We independently process camera images within the proposed comprehensive image processing pipeline. Specifically, first, we transform the camera images to Bird's-Eye View (BEV) Polar domain and extract the corresponding features with our camera encoder-decoder architecture. The resultant feature maps are fused with Range-Azimuth (RA) features, recovered from the RD spectrum input from the radar decoder to perform object detection. We evaluate our fusion strategy with other existing methods not only in terms of accuracy but also on computational complexity metrics on RADIal dataset.