CVRR dataset for trajectory clustering|轨迹聚类数据集|计算机视觉数据集

Six trajectory clustering datasets (Morris and Trivedi, 2009) are provided for benchmarking trajectory clustering algorithms. These datasets cover a diverse range of scenes to enable thorough evaluation of different algorithms. The datasets include three simulated highway scenes, a real highway scene, a simulated intersection scene, and an indoor omnidirectional camera scene.These datasets originate from the work by Morris and Trivedi (2009), described in their paper "Learning trajectory patterns by clustering: Experimental studies and comparative evaluation," presented at the 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops. The datasets provide a valuable resource for the trajectory clustering community, enabling comparative evaluation and benchmarking of various clustering algorithms. The datasets can be accessed through the following DOI link: https://doi.org/10.1109/CVPR.2009.5206559.