SUTrack: Towards Simple and Unified Single Object Tracking

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Xin, Kang, Ben, Geng, Wanting, Zhu, Jiawen, Liu, Yi, Wang, Dong, Lu, Huchuan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917879704715264
author Chen, Xin
Kang, Ben
Geng, Wanting
Zhu, Jiawen
Liu, Yi
Wang, Dong
Lu, Huchuan
author_facet Chen, Xin
Kang, Ben
Geng, Wanting
Zhu, Jiawen
Liu, Yi
Wang, Dong
Lu, Huchuan
contents In this paper, we propose a simple yet unified single object tracking (SOT) framework, dubbed SUTrack. It consolidates five SOT tasks (RGB-based, RGB-Depth, RGB-Thermal, RGB-Event, RGB-Language Tracking) into a unified model trained in a single session. Due to the distinct nature of the data, current methods typically design individual architectures and train separate models for each task. This fragmentation results in redundant training processes, repetitive technological innovations, and limited cross-modal knowledge sharing. In contrast, SUTrack demonstrates that a single model with a unified input representation can effectively handle various common SOT tasks, eliminating the need for task-specific designs and separate training sessions. Additionally, we introduce a task-recognition auxiliary training strategy and a soft token type embedding to further enhance SUTrack's performance with minimal overhead. Experiments show that SUTrack outperforms previous task-specific counterparts across 11 datasets spanning five SOT tasks. Moreover, we provide a range of models catering edge devices as well as high-performance GPUs, striking a good trade-off between speed and accuracy. We hope SUTrack could serve as a strong foundation for further compelling research into unified tracking models. Code and models are available at github.com/chenxin-dlut/SUTrack.
format Preprint
id arxiv_https___arxiv_org_abs_2412_19138
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SUTrack: Towards Simple and Unified Single Object Tracking
Chen, Xin
Kang, Ben
Geng, Wanting
Zhu, Jiawen
Liu, Yi
Wang, Dong
Lu, Huchuan
Computer Vision and Pattern Recognition
Machine Learning
In this paper, we propose a simple yet unified single object tracking (SOT) framework, dubbed SUTrack. It consolidates five SOT tasks (RGB-based, RGB-Depth, RGB-Thermal, RGB-Event, RGB-Language Tracking) into a unified model trained in a single session. Due to the distinct nature of the data, current methods typically design individual architectures and train separate models for each task. This fragmentation results in redundant training processes, repetitive technological innovations, and limited cross-modal knowledge sharing. In contrast, SUTrack demonstrates that a single model with a unified input representation can effectively handle various common SOT tasks, eliminating the need for task-specific designs and separate training sessions. Additionally, we introduce a task-recognition auxiliary training strategy and a soft token type embedding to further enhance SUTrack's performance with minimal overhead. Experiments show that SUTrack outperforms previous task-specific counterparts across 11 datasets spanning five SOT tasks. Moreover, we provide a range of models catering edge devices as well as high-performance GPUs, striking a good trade-off between speed and accuracy. We hope SUTrack could serve as a strong foundation for further compelling research into unified tracking models. Code and models are available at github.com/chenxin-dlut/SUTrack.
title SUTrack: Towards Simple and Unified Single Object Tracking
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2412.19138