{"title":"A review of object tracking based on deep learning","authors":"Guochen Zhao , Fanyong Meng , Chengzhuan Yang , Hui Wei , Dawei Zhang , Zhonglong Zheng","doi":"10.1016/j.neucom.2025.130988","DOIUrl":null,"url":null,"abstract":"<div><div>The rapid advancement of deep learning has led to a surge in the development of object-tracking algorithms. Given the diverse objectives, backbone networks, and application methodologies, this study aims to integrate the prevalent tracking approaches comprehensively. We propose a systematic classification scheme based on application scenarios and primary methods, accompanied by a thorough analysis and concise summaries of each category. This approach provides a broader coverage of tracking techniques, facilitating a quicker understanding of the domain for novice researchers. In addition, we present standardized evaluation metrics and widely used datasets, including cross-method performance comparisons of selected algorithms on identical benchmarks to enhance the reader’s contextual understanding. Finally, we offer a critical assessment of current limitations, practical recommendations, and forward-looking perspectives to guide future research directions.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 130988"},"PeriodicalIF":6.5000,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neurocomputing","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0925231225016601","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
The rapid advancement of deep learning has led to a surge in the development of object-tracking algorithms. Given the diverse objectives, backbone networks, and application methodologies, this study aims to integrate the prevalent tracking approaches comprehensively. We propose a systematic classification scheme based on application scenarios and primary methods, accompanied by a thorough analysis and concise summaries of each category. This approach provides a broader coverage of tracking techniques, facilitating a quicker understanding of the domain for novice researchers. In addition, we present standardized evaluation metrics and widely used datasets, including cross-method performance comparisons of selected algorithms on identical benchmarks to enhance the reader’s contextual understanding. Finally, we offer a critical assessment of current limitations, practical recommendations, and forward-looking perspectives to guide future research directions.
期刊介绍:
Neurocomputing publishes articles describing recent fundamental contributions in the field of neurocomputing. Neurocomputing theory, practice and applications are the essential topics being covered.