首页|Investigators at New York University (NYU) Report Findings in Robotics and Autom ation (Salsa: Swift Adaptive Lightweight Selfattention for Enhanced Lidar Place Recognition)
Investigators at New York University (NYU) Report Findings in Robotics and Autom ation (Salsa: Swift Adaptive Lightweight Selfattention for Enhanced Lidar Place Recognition)
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-Investigators publish new report on Ro botics-Robotics and Automation. According to news originating from Brooklyn, N ew York, by NewsRx correspondents, research stated, "Large-scale LiDAR mappings and localization leverage place recognition techniques to mitigate odometry drif ts, ensuring accurate mapping. These techniques utilize scene representations fr om LiDAR point clouds to identify previously visited sites within a database." Funders for this research include ARO, New York University Abu Dhabi (NYUAD) Cen ter for Artificial Intelligence and Robotics (CAIR)-Tamkeen through NYUAD Rese arch Institute Award. Our news journalists obtained a quote from the research from New York University (NYU), "Local descriptors, assigned to each point within a point cloud, are agg regated to form a scene representation for the point cloud. These descriptors ar e also used to re-rank the retrieved point clouds based on geometric fitness sco res. We propose SALSA, a novel, lightweight, and efficient framework for LiDAR p lace recognition. It consists of a Sphereformer backbone that uses radial window attention to enable information aggregation for sparse distant points, an adapt ive self-attention layer to pool local descriptors into tokens, and a multi-laye r-perceptron Mixer layer for aggregating the tokens to generate a scene descript or."
BrooklynNew YorkUnited StatesNorth and Central AmericaRobotics and AutomationRoboticsNew York University (NY U)
NETL
NSTL
