首页|New Robotics and Mechatronics Research from University of Tokyo Described (Data Fusion for Sparse Semantic Localization Based on Object Detection)
New Robotics and Mechatronics Research from University of Tokyo Described (Data Fusion for Sparse Semantic Localization Based on Object Detection)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News - A new study on robotics and mechatroni cs is now available. According to newsoriginating from Tokyo, Japan, by NewsRx correspondents, research stated, “Semantic information hasstarted to be used in localization methods to introduce a non-geometric distinction in the environmen t.”Our news editors obtained a quote from the research from University of Tokyo: “H owever, efficientways to integrate this information remain a question. We propo se an approach for fusing data from different object classes by analyzing the po sterior for each object class to improve robustness and accuracyfor self-locali zation. Our system uses the bearing angle to the objects’ center and objects’ cl ass namesas sensor model input to localize the user on a 2D annotated map consi sting of objects’ class names andcenter coordinates. Sensor model input is obta ined by an object detector on equirectangular images ofa 360° field of view cam era. As object detection performance varies based on location and object class,different object classes generate different likelihoods.”
University of TokyoTokyoJapanAsiaMechatronicsRobotics
NETL
NSTL
