Detect3D-Lidar-DepthCamera

基于聚类方法的2D到3D目标检测方法包含密度聚类和半监督PCKMeans半监督聚类 Yolov5目标检测方法--TensorRT加速支持多线激光雷达+RGB相机、深度相机两种方式 ROS下编译运行模型选择位于src/detect.cpp中可更换自己的目标检测模型但需要在TensorRT下支持

std::string weight="/home/ax/code/Yolov5_trt/engine/";
std::string classname_path="/home/ax/code/Yolov5_trt/engine/coco.names";

模式选择位位于include/include.h中可选雷达或深度相机模式

#define DepthEnable 1
#define LidarEnable 0

演示视频

在Lego-LOAM下机器人运行效果

待更新～

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
include		include
launch		launch
msg		msg
src		src
CMakeLists.txt		CMakeLists.txt
README.md		README.md
package.xml		package.xml