• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops >Temporal Alignment Improves Feature Quality: An Experiment on Activity Recognition with Accelerometer Data
【24h】

Temporal Alignment Improves Feature Quality: An Experiment on Activity Recognition with Accelerometer Data

机译：时间对准改善了特征质量：加速度计数据的活动识别实验

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Activity recognition has been receiving significant attention from a variety of research areas such as human performance enhancement, health promotion, and human computer interaction. However, recognizing activities from accelerometer data still remains a challenging problem due to sensitivity to sampling rates, misalignment of data, and increased variability in activities among clinically relevant populations. In order to solve these issues, we adopt methods from functional analysis, which consider non-elastic rate variations in movement. The overall framework factors out temporal variability within activity classes, before leveraging robust machine learning pipelines for a given end-use. The proposed approach has been evaluated on 7 classes of everyday activities with 50 subjects. The results indicate that proposed approach achieves improved performance with the improvements observed in separating similar classes that differ in temporal rates, and also demonstrate higher robustness to change in window lengths. These results suggest that temporal alignment should be considered a core part of activity recognition pipelines.
机译：活动识别一直受到各种研究领域的重大关注，如人类性能增强，健康促进和人机互动。然而，由于对采样率，对数据的不对准的敏感性，以及临床相关人群的活动的增加，识别来自加速度计数据的活动仍然是一个具有挑战性的问题。为了解决这些问题，我们采用了功能分析的方法，这考虑了运动的非弹性速率变化。在利用强大的机器学习管道以获得给定的最终使用之前，整体框架因素突出了活动类内的时间变异性。拟议的方法已在7个日常活动中评估了50个科目的评估。结果表明，提出的方法实现了改进的性能，在分离时间率不同的类似类别方面观察到的改进，并且还表明窗口长度的变化更高的鲁棒性。这些结果表明，时间对准应被视为活动识别管道的核心部分。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号