Skip to content

roi_cluster_fusion#

Purpose#

roi_cluster_fusion is a package for filtering clusters that are less likely to be objects and overwriting labels of clusters with that of Region Of Interests (ROIs) by a 2D object detector.

Inner-workings / Algorithms#

The clusters are projected onto image planes, and then if the ROIs of clusters and ROIs by a detector are overlapped, the labels of clusters are overwritten with that of ROIs by detector. Intersection over Union (IoU) is used to determine if there are overlaps between them.

roi_cluster_fusion_image

Inputs / Outputs#

Input#

Name Type Description
clusters tier4_perception_msgs::msg::DetectedObjectsWithFeature clustered pointcloud
input/camera_infoID sensor_msgs::msg::CameraInfo camera information to project 3d points onto image planes, ID is between 0 and 7
input/roisID tier4_perception_msgs::msg::DetectedObjectsWithFeature ROIs from each image, ID is between 0 and 7
input/image_rawID sensor_msgs::msg::Image images for visualization, ID is between 0 and 7

Output#

Name Type Description
output/image_rawID sensor_msgs::msg::Image images for visualization, ID is between 0 and 7

Parameters#

Core Parameters#

Name Type Description
use_iou_x bool calculate IoU only along x-axis
use_iou_y bool calculate IoU only along y-axis
use_iou bool calculate IoU both along x-axis and y-axis
use_cluster_semantic_type bool if false, the labels of clusters are overwritten by UNKNOWN before fusion
iou_threshold float the IoU threshold to overwrite a label of clusters with a label of roi
rois_number int the number of input rois

Assumptions / Known limits#

(Optional) Error detection and handling#

(Optional) Performance characterization#

(Optional) Future extensions / Unimplemented parts#

Back to top