Compare the tracker and unknown objects, and determine that those with large recall and small precision are under segmented objects. Policy for dealing with under segmentation # Shape fitting using the tracker information such as angle and size as reference information. Merge the unknown objects in the tracker as a single object. Policy for dealing with over segmentation # We need to take measures for under segmentation and over segmentation. Simply looking at the overlap between the unknown object and the tracker does not work. Shape fitting using euclidean clustering and other methods has a problem called under segmentation and over segmentation. The detection by tracker receives an unknown object containing a point cloud and a tracker, where the unknown object is mainly shape-fitted using euclidean clustering. The unknown object is optimized to fit the size of the tracker so that it can continue to be detected. The detection by tracker takes as input an unknown object containing a cluster of points and a tracker. This package feeds back the tracked objects to the detection module to keep it stable and keep detecting objects. Pointcloud_preprocessor : occupancy_grid_map_outlier_filter The `freespace planning algorithms` package The `traffic_light_map_based_detector` Package 
(Optional) Future extensions / Unimplemented parts Policy for dealing with under segmentation
Policy for dealing with over segmentation
0 kommentar(er)
