Saved in:
Bibliographic Details
Main Authors: Alikhanov, Jumabek, Obidov, Dilshod, Kim, Hakil
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.04187
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913524590051328
author Alikhanov, Jumabek
Obidov, Dilshod
Kim, Hakil
author_facet Alikhanov, Jumabek
Obidov, Dilshod
Kim, Hakil
contents The Lightweight Integrated Tracking-Feature Extraction (LITE) paradigm is introduced as a novel multi-object tracking (MOT) approach. It enhances ReID-based trackers by eliminating inference, pre-processing, post-processing, and ReID model training costs. LITE uses real-time appearance features without compromising speed. By integrating appearance feature extraction directly into the tracking pipeline using standard CNN-based detectors such as YOLOv8m, LITE demonstrates significant performance improvements. The simplest implementation of LITE on top of classic DeepSORT achieves a HOTA score of 43.03% at 28.3 FPS on the MOT17 benchmark, making it twice as fast as DeepSORT on MOT17 and four times faster on the more crowded MOT20 dataset, while maintaining similar accuracy. Additionally, a new evaluation framework for tracking-by-detection approaches reveals that conventional trackers like DeepSORT remain competitive with modern state-of-the-art trackers when evaluated under fair conditions. The code will be available post-publication at https://github.com/Jumabek/LITE.
format Preprint
id arxiv_https___arxiv_org_abs_2409_04187
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LITE: A Paradigm Shift in Multi-Object Tracking with Efficient ReID Feature Integration
Alikhanov, Jumabek
Obidov, Dilshod
Kim, Hakil
Computer Vision and Pattern Recognition
The Lightweight Integrated Tracking-Feature Extraction (LITE) paradigm is introduced as a novel multi-object tracking (MOT) approach. It enhances ReID-based trackers by eliminating inference, pre-processing, post-processing, and ReID model training costs. LITE uses real-time appearance features without compromising speed. By integrating appearance feature extraction directly into the tracking pipeline using standard CNN-based detectors such as YOLOv8m, LITE demonstrates significant performance improvements. The simplest implementation of LITE on top of classic DeepSORT achieves a HOTA score of 43.03% at 28.3 FPS on the MOT17 benchmark, making it twice as fast as DeepSORT on MOT17 and four times faster on the more crowded MOT20 dataset, while maintaining similar accuracy. Additionally, a new evaluation framework for tracking-by-detection approaches reveals that conventional trackers like DeepSORT remain competitive with modern state-of-the-art trackers when evaluated under fair conditions. The code will be available post-publication at https://github.com/Jumabek/LITE.
title LITE: A Paradigm Shift in Multi-Object Tracking with Efficient ReID Feature Integration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.04187