CenterFusion: Center-based Radar and Camera Fusion for 3D Object Detection

TL;DR

CenterFusion is a novel middle-fusion approach that combines radar and camera data for improved 3D object detection in autonomous vehicles, enhancing accuracy and velocity estimation without extra temporal data.

Contribution

It introduces a new frustum-based data association method and demonstrates significant improvements over camera-only methods on the nuScenes dataset.

Findings

Over 12% improvement in nuScenes Detection Score (NDS)

Significant enhancement in velocity estimation accuracy

Effective fusion of radar and camera data without temporal info

Abstract

The perception system in autonomous vehicles is responsible for detecting and tracking the surrounding objects. This is usually done by taking advantage of several sensing modalities to increase robustness and accuracy, which makes sensor fusion a crucial part of the perception system. In this paper, we focus on the problem of radar and camera sensor fusion and propose a middle-fusion approach to exploit both radar and camera data for 3D object detection. Our approach, called CenterFusion, first uses a center point detection network to detect objects by identifying their center points on the image. It then solves the key data association problem using a novel frustum-based method to associate the radar detections to their corresponding object's center point. The associated radar detections are used to generate radar-based feature maps to complement the image features, and regress to object properties such as depth, rotation and velocity. We evaluate CenterFusion on the challenging nuScenes dataset, where it improves the overall nuScenes Detection Score (NDS) of the state-of-the-art camera-based algorithm by more than 12%. We further show that CenterFusion significantly improves the velocity estimation accuracy without using any additional temporal information. The code is available at https://github.com/mrnabati/CenterFusion .