Scholarlink

<p>Gait recognition enables non-intrusive identification of individuals through their walking patterns, eliminating the need for active cooperation. It has gained popularity due to its remote applicability and ability to work with low-resolution videos. This study explores the effectiveness of Scale-Invariant Feature Transform and Speeded-Up Robust Features for gait recognition. Decision Tree and Random Forest classifiers are employed to evaluate the recognition performance of these features, achieving accuracies of 85.20% and 87.80%, respectively, on the CASIA-A dataset.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

MLforGait: Using Machine Learning Models for Automatic Gait Recognition Without Subject Cooperation

  • 期刊论文

  • 发表日期：

    2025年6月13日

查看全文
  • Muhammad Irsyad Abdullah,
  • Gadug Sudhamsu,
  • Pooja Rani,
  • Jayant Jagtap,
  • Mandeep Kaur Chohan,
  • M. Janaki Ramudu,
  • Devendra Singh,
  • Ahmed Alkhayyat

摘要

Gait recognition enables non-intrusive identification of individuals through their walking patterns, eliminating the need for active cooperation. It has gained popularity due to its remote applicability and ability to work with low-resolution videos. This study explores the effectiveness of Scale-Invariant Feature Transform and Speeded-Up Robust Features for gait recognition. Decision Tree and Random Forest classifiers are employed to evaluate the recognition performance of these features, achieving accuracies of 85.20% and 87.80%, respectively, on the CASIA-A dataset.