HARadNet: Anchor-free target detection for radar point clouds using hierarchical attention and multi-task learning

Dubey A, Santra A, Fuchs J, Lübke M, Weigel R, Lurz F (2022)


Publication Language: English

Publication Type: Journal article, Original article

Publication year: 2022

Journal

Pages Range: 100275

Article Number: 100275

DOI: 10.1016/j.mlwa.2022.100275

Open Access Link: https://doi.org/10.1016/j.mlwa.2022.100275

Abstract

Target localization and classification from radar point clouds is a challenging task due to the inherently sparse nature of the data with highly non-uniform target distribution. This work presents HARadNet, a novel attention based anchor free target detection and classification network architecture in a multi-task learning framework for radar point clouds data. A direction field vector is used as motion modality to achieve attention inside the network. The attention operates at different hierarchy of the feature abstraction layer with each point sampled according to a conditional direction field vector, allowing the network to exploit and learn a joint feature representation and correlation to its neighborhood. This leads to a significant improvement in the performance of the classification. Additionally, a parameter-free target localization is proposed using Bayesian sampling conditioned on a pre-trained direction field vector. The extensive evaluation on a public radar dataset shows an substantial increase in localization and classification performance.

Authors with CRIS profile

Related research project(s)

Involved external institutions

How to cite

APA:

Dubey, A., Santra, A., Fuchs, J., Lübke, M., Weigel, R., & Lurz, F. (2022). HARadNet: Anchor-free target detection for radar point clouds using hierarchical attention and multi-task learning. Machine Learning with Applications, 100275. https://dx.doi.org/10.1016/j.mlwa.2022.100275

MLA:

Dubey, Anand, et al. "HARadNet: Anchor-free target detection for radar point clouds using hierarchical attention and multi-task learning." Machine Learning with Applications (2022): 100275.

BibTeX: Download