Excerpt of Auritus: An Open-Source Optimization Toolkit for Training and Development of Human Movement Models and Filters Using Earables
Excerpt of Auritus: An Open-Source Optimization Toolkit for Training and Development of Human Movement Models and Filters Using Earables
复制标题
Auritus 摘录：用于使用 Earables 训练和开发人体运动模型和过滤器的开源优化工具包
DOI：
10.1145/3544793.3563423
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Srivastava, Mani
中科院分区：
文献类型：
--
作者：
Saha, Swapnil Sayan;Sandha, Sandeep Singh;Pei, Siyou;Jain, Vivek;Wang, Ziqi;Li, Yuchen;Sarker, Ankur;Srivastava, Mani
Auritus 是一个可扩展的开源优化工具包，旨在增强和复制耳戴式应用程序。 Auritus 有两个主要功能。首先，Auritus 处理数据收集、预处理和标记任务，以使用图形工具创建定制的可听数据集。该系统包括一个开源数据集，其中包含 243 万个与头部和全身运动相关的惯性样本，其中包括来自 45 名志愿者的 34 个头部姿势和 9 个活动。其次，Auritus 提供紧密集成的硬件在环 (HIL) 优化器和 TinyML 接口，用于开发用于活动检测的轻量级实时机器学习 (ML) 模型和用于头部姿势跟踪的过滤器。 Auritus 使用小至 6-13 kB 的实时模型以 91% 的留一测试准确率（98% 的测试准确率）识别活动。我们的模型比最先进的模型小 98-740 倍，准确度提高 3-6%。我们还使用 20kB 滤波器以低至 5 度的绝对误差估计头部姿势，与现有技术相比，精度提高了 1.6 倍。 Auritus 可以在 https://github.com/nesl/auritus 上找到。
Auritus is an extendable and open-source optimization toolkit designed to enhance and replicate earable applications. Auritus serves two primary functions. Firstly, Auritus handles data collection, pre-processing, and labeling tasks for creating customized earable datasets using graphical tools. The system includes an open-source dataset with 2.43 million inertial samples related to head and full-body movements, consisting of 34 head poses and 9 activities from 45 volunteers. Secondly, Auritus provides a tightly-integrated hardware-in-the-loop (HIL) optimizer and TinyML interface to develop lightweight and real-time machine-learning (ML) models for activity detection and filters for head-pose tracking. Auritus recognizes activities with 91% leave 1-out test accuracy (98% test accuracy) using real-time models as small as 6-13 kB. Our models are 98-740 × smaller and 3-6% more accurate over the state-of-the-art. We also estimate head pose with absolute errors as low as 5 degrees using 20kB filters, achieving up to 1.6 × precision improvement over existing techniques. Auritus is available at https://github.com/nesl/auritus.