FADet: A Multi-sensor 3D Object Detection Network based on Local Featured Attention

TL;DR

FADet is a multi-sensor 3D object detection network that effectively fuses camera, LiDAR, and radar data using local featured attention modules, achieving state-of-the-art results on NuScenes dataset.

Contribution

The paper introduces local featured attention modules tailored for each sensor type, enhancing multi-sensor fusion for 3D detection in autonomous driving.

Findings

Achieves 71.8% NDS and 69.0% mAP on LiDAR-camera detection

Achieves 51.7% NDS and 40.3% mAP on radar-camera detection

Effective in long-tail and complex scenes

Abstract

Camera, LiDAR and radar are common perception sensors for autonomous driving tasks. Robust prediction of 3D object detection is optimally based on the fusion of these sensors. To exploit their abilities wisely remains a challenge because each of these sensors has its own characteristics. In this paper, we propose FADet, a multi-sensor 3D detection network, which specifically studies the characteristics of different sensors based on our local featured attention modules. For camera images, we propose dual-attention-based sub-module. For LiDAR point clouds, triple-attention-based sub-module is utilized while mixed-attention-based sub-module is applied for features of radar points. With local featured attention sub-modules, our FADet has effective detection results in long-tail and complex scenes from camera, LiDAR and radar input. On NuScenes validation dataset, FADet achieves state-of-the-art performance on LiDAR-camera object detection tasks with 71.8% NDS and 69.0% mAP, at the same time, on radar-camera object detection tasks with 51.7% NDS and 40.3% mAP. Code will be released at https://github.com/ZionGo6/FADet.