ExoNet Database: Wearable Camera Images of Human Locomotion Environments

Advances in artificial intelligence and robotic vision have enabled researchers to develop environment recognition systems for lower-limb exoskeletons and prostheses. However, insufficient and private training datasets have impeded the widespread development and dissemination of image classification algorithms. To address these shortcomings, we developed the ExoNet Database, the first open-source database of high-resolution wearable RGB camera images of human locomotion environments. Using a lightweight wearable smartphone camera system, over 5.4 million images of outdoor and indoor real-world locomotion environments were collected throughout summer, autumn, and winter seasons. Approximately 900,000 images of the ExoNet Database were human-annotated using a 12-classs hierarchical image classification architecture. Im¬ages were uploaded to IEEE DataPort and are publicly available for download. The ExoNet Database provides an unprecedented communal platform for training, developing, and comparing next-generation image classification algorithms for control of lower-limb exoskeletons and prostheses.

人工智能与机器视觉领域的技术进展，已助力研究者开发出适用于下肢外骨骼与假肢的环境识别系统。然而，训练数据集的匮乏与私有属性，阻碍了图像分类算法的广泛开发与推广应用。为解决上述缺陷，我们构建了ExoNet数据集（ExoNet Database）——首个面向人类行走环境的高分辨率可穿戴RGB相机图像开源数据库。本研究采用轻量化可穿戴智能手机相机系统，在夏、秋、冬三季采集了超过540万张真实室内外人类行走环境的图像。ExoNet数据集中约90万张图像，通过12类分层图像分类架构完成了人工标注。所有图像已上传至IEEE DataPort平台，可公开下载获取。本数据集为面向下肢外骨骼与假肢控制的下一代图像分类算法的训练、开发与对比，提供了前所未有的共享研究平台。

搜集汇总

背景与挑战

背景概述

ExoNet是目前最大的开源可穿戴摄像头图像数据集，包含560万张RGB图像和92.3万张人工标注图像，覆盖不同季节的室内外行走环境，采用12类分层标签架构，为深度学习模型开发提供平台。