首页|Data on Robotics and Automation Reported by Researchers at University of British Columbia (Multi-modal Streaming 3d Object Detection)
Data on Robotics and Automation Reported by Researchers at University of British Columbia (Multi-modal Streaming 3d 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 Daily News Daily News-Current study results on Robotics - Ro botics and Automation have been published.According to news reporting from Vanc ouver,Canada,by NewsRx journalists,research stated,"Modern autonomous vehicl es rely heavily on mechanical LiDARs for perception.Current perception methods generally require $360 <.> \circ$ point clouds,collected sequentia lly as the LiDAR scans the azimuth and acquires consecutive wedge-shaped slices." Financial support for this research came from Natural Sciences and Engineering R esearch Council of Canada (NSERC).The news correspondents obtained a quote from the research from the University o f British Columbia,"The acquisition latency of a full scan ($\ sim\!\text{100} \;\text{ms} $) may lead to outdated perception which is detrimental to safe ope ration.Recent streaming perception works proposed directly processing LiDAR sli ces and compensating for the narrow field of view (FOV) of a slice by reusing fe atures from preceding slices.These works,however,are all based on a single mo dality and require past information which may be outdated.Meanwhile,images fro m high-frequency cameras can support streaming models as they provide a larger F oV compared to a LiDAR slice.However,this difference in FoV complicates sensor fusion.We propose an innovative camera-LiDAR streaming 3D object detection fra mework that uses camera images instead of past LiDAR slices to provide an up-to- date,dense,and wide context for streaming perception.The proposed method outp erforms prior streaming models and powerful full-scan baselines on the challengi ng NuScenes benchmark in detection accuracy and end-to-end runtime."
VancouverCanadaNorth and Central Ame ricaRobotics and AutomationRoboticsUniversity of British Columbia
NETL
NSTL
