ExoNet Database: Open-Source Wearable Camera Images of Human Locomotion Environments
收藏资源简介：
Recent advances in robotic vision and artificial intelligence have allowed researchers to develop environment recognition systems for lower-limb exoskeletons and prostheses. However, insufficient and private training databases have impeded the widespread development and dissemination of image classification algorithms for environment recognition. To address these shortcomings, we have developed “ExoNet”, the first open-source large-scale hierarchical database of high-resolution wearable camera images of human locomotion environments. Unparalleled in both scale and diversity, ExoNet comprises over 5.6 million images of different indoor and outdoor real-world walking environments, collected using a lightweight wearable smartphone camera system throughout summer, autumn, and winter seasons. Approximately 940,000 images in ExoNet were human-annotated using a 12-class hierarchical classification architecture. Available publicly through IEEE DataPort, ExoNet offers an unprecedented communal platform for training, developing, and comparing image classification algorithms for next-generation environment recognition systems. Beyond the control of robotic lower-limb exoskeletons and prostheses, applications of ExoNet extend to humanoid and autonomous legged robotics.
机器人视觉与人工智能领域的新近进展，已助力研究者研发出适用于下肢外骨骼与假肢的环境识别系统。然而，规模有限且受隐私约束的训练数据集，阻碍了环境识别用图像分类算法的大规模开发与推广应用。为解决上述不足，本研究构建了ExoNet——首个开源大规模层级化数据集，收录人类运动环境下的高分辨率可穿戴相机图像。该数据集在规模与多样性上均无可匹敌，涵盖超560万张涵盖不同室内外真实步行环境的图像，采集时段覆盖夏、秋、冬三季，采集设备为轻量化可穿戴智能手机相机系统。其中约94万张图像已通过12类层级分类架构完成人工标注。ExoNet可通过IEEE DataPort公开获取，为下一代环境识别系统的图像分类算法训练、开发与对比提供了前所未有的公共研究平台。除应用于下肢外骨骼与假肢的机器人控制系统外，ExoNet还可拓展至人形机器人与自主式足式机器人领域。
