SIRA: Scalable Inter-frame Relation and Association for Radar Perception
Preprint
Abstract
Conventional radar feature extraction faces limitations due to low spatial resolution, noise, multipath reflection, the presence of ghost targets, and motion blur. Such limitations can be exacerbated by nonlinear object motion, particularly from an ego-centric viewpoint. It becomes evident that to address these challenges, the key lies in exploiting temporal feature relation over an extended horizon and enforcing spatial motion consistency for effective association. To this end, this paper proposes SIRA (Scalable Inter-frame Relation and Association) with two designs. First, inspired by Swin Transformer, we introduce extended temporal relation, generalizing the existing temporal relation layer from two consecutive frames to multiple inter-frames with temporally regrouped window attention for scalability. Second, we propose motion consistency track with the concept of a pseudo-tracklet generated from observational data for better trajectory prediction and subsequent object association. Our approach achieves 58.11 mAP@0.5 for oriented object detection and 47.79 MOTA for multiple object tracking on the Radiate dataset, surpassing previous state-of-the-art by a margin of +4.11 mAP@0.5 and +9.94 MOTA, respectively.Keywords
All Related Versions
- Version 1, 2024-11-04, ArXiv
- Published version: , 15024.
This publication has 16 references indexed in Scilit:
- Sensing and Machine Learning for Automotive Perception: A ReviewIEEE Sensors Journal, 2023
- Multi-Class Road User Detection With 3+1D Radar in the View-of-Delft DatasetIEEE Robotics and Automation Letters, 2022
- FairMOT: On the Fairness of Detection and Re-identification in Multiple Object TrackingInternational Journal of Computer Vision, 2021
- Learning-Based Extended Object Tracking Using Hierarchical Truncation Measurement Model With Automotive RadarIEEE Journal of Selected Topics in Signal Processing, 2021
- RAMP-CNN: A Novel Neural Network for Enhanced Automotive Radar Object RecognitionIEEE Sensors Journal, 2020
- HOTA: A Higher Order Metric for Evaluating Multi-object TrackingInternational Journal of Computer Vision, 2020
- The Rise of Radar for Autonomous Vehicles: Signal Processing Solutions and Future Research DirectionsIEEE Signal Processing Magazine, 2019
- Extended Object Tracking with Random Hypersurface ModelsIEEE Transactions on Aerospace and Electronic Systems, 2014
- Bayesian approach to extended object and cluster tracking using random matricesIEEE Transactions on Aerospace and Electronic Systems, 2008
- Particle filters for positioning, navigation, and trackingIEEE Transactions on Signal Processing, 2002