Online Multiple Object Tracking Using Spatial Pyramid Pooling Hashing and Image Retrieval for Autonomous Driving

Multiple object tracking (MOT) is a fundamental issue and has attracted considerable attention in the autonomous driving community. This paper presents a novel MOT framework for autonomous driving. The framework consists of two stages of object representation and data association. In the stage of ob...

Full description

Saved in:
Bibliographic Details
Published inMachines (Basel) Vol. 10; no. 8; p. 668
Main Authors Wei, Hongjian, Huang, Yingping
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.08.2022
Subjects
Affinity
Automobile driving
data association
Image processing
Image retrieval
Innovations
Methods
Motor vehicle driving
multiple object tracking
Multiple target tracking
Neural networks
Object recognition
object representation
Occlusion
Optical flow (image analysis)
Representations
Semantics
spatial pyramid pooling hashing
Technology application
Tracking systems
Vision systems
China
Online AccessGet full text

Cover

More Information
Summary:Multiple object tracking (MOT) is a fundamental issue and has attracted considerable attention in the autonomous driving community. This paper presents a novel MOT framework for autonomous driving. The framework consists of two stages of object representation and data association. In the stage of object representation, we employ appearance, motion, and position features to characterize objects. We design a spatial pyramidal pooling hash network (SPPHNet) to generate the appearance features. Multiple-level representative features in the SPPHNet are mapped into a similarity-preserving binary space, called hash features. The hash features retain the visual discriminability of high-dimensional features and are beneficial for computational efficiency. For data association, a two-tier data association scheme is designed to address the occlusion issue, consisting of an affinity cost model and a hash-based image retrieval model. The affinity cost model accommodates the hash features, disparity, and optical flow as the first tier of data association. The hash-based image retrieval model exploits the hash features and adopts image retrieval technology to handle reappearing objects as the second tier of data association. Experiments on the KITTI public benchmark dataset and our campus scenario sequences show that our method has superior tracking performance to the state-of-the-art vision-based MOT methods.
ISSN:2075-1702
2075-1702
DOI:10.3390/machines10080668