查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Robotics is the subjec t of a report. According to news originating from Toronto, Canada, by NewsRx cor respondents, research stated, "Human-robot walking with prosthetic legs and exos keletons, especially over complex terrains, such as stairs, remains a significan t challenge. Egocentric vision has the unique potential to detect the walking en vironment prior to physical interactions, which can improve transitions to and f rom stairs." Funders for this research include AGE-WELL, Vector Institute, The Schroeder Inst itute for Brain Innovation and Recovery. Our news journalists obtained a quote from the research from the University of T oronto, "This motivated us to develop the StairNet initiative to support the dev elopment of new deep learning models for visual perception of real-world stair e nvironments. In this study, we present a comprehensive overview of the StairNet initiative and key research to date. First, we summarize the development of our large-scale data set with over 515,000 manually labeled images. We then provide a summary and detailed comparison of the performances achieved with different al gorithms (i.e., 2D and 3D CNN, hybrid CNN and LSTM, and ViT networks), training methods (i.e., supervised learning with and without temporal data, and semi-supe rvised learning with unlabeled images), and deployment methods (i.e., mobile and embedded computing), using the StairNet data set. Finally, we discuss the chall enges and future directions. To date, our StairNet models have consistently achi eved high classification accuracy (i.e., up to 98.8%) with differen t designs, offering trade-offs between model accuracy and size. When deployed on mobile devices with GPU and NPU accelerators, our deep learning models achieved inference speeds up to 2.8 ms. In comparison, when deployed on our custom-desig ned CPU-powered smart glasses, our models yielded slower inference speeds of 1.5 s, presenting a trade-off between human-centered design and performance. Overal l, the results of numerous experiments presented herein provide consistent evide nce that StairNet can be an effective platform to develop and study new deep lea rning models for visual perception of human-robot walking environments, with an emphasis on stair recognition."
NETL
NSTL