leofansq

Tools_RosBag2KITTI

Conversion from ROSBAG (.bag) to image (.png) and points cloud (.bin), including ROSBAG decoding, pcd2bin and file directory extraction.

Tools_Merge_Image_PointCloud

Project the PointCloud to the image & Generate the LiDAR PointCloud with color.

SCF-Net

[CVPR2021] SCF-Net: Learning Spatial Contextual Features for Large-Scale Point Cloud Segmentation

AI01001H-PatternRecognition

UCAS 081100M01001H 模式识别: Python实现实例与实验

SIFT

基于C++的基本库实现了SIFT特征提取与匹配, 无需调用如OpenCV的库函数. 从实现细节的角度介绍了SIFT算法流程. 代码很大程度地基于ezSIFT代码, 用于自学和理解SIFT算法.

CBR

[IROS2023] Calibration-free BEV Representation for Infrastructure Perception

Reinforcement_Learning_Curling

基于强化学习(RL)的冰壶游戏实例; 梯度下降的Sarsa(lambda) + 非均匀径向基特征表示

ROS_Pure_Pursuit

Pure pursuit in pure ROS (Melodic)

AI01002H-ImageProcessing

UCAS 081100M01002H 图像处理与分析: Python实现实例与实验

CPCL

[IEEE TIP] Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation

Tools_make_planes

AVOD needs the planes file to provide ground plane information, but the official planes generation tool has not yet been provided, which brings great difficulty to the test work. This project is used to generate planes files especially for AVOD testing.

FII-CenterNet

[IEEE TVT] FII-CenterNet: an anchor-free detector with foreground attention for traffic object detection

ROS_NDT_Mapping

NDT Mapping in pure ROS (Melodic)

Lane_Follower

Lane follower: Lane-detector (OpenCV) + Object-detector (YOLO5) + CAN-bus

ROS_NDT_Localizer

NDT Localizer in pure ROS (Melodic)

SpiderMesh

SpiderMesh: Spatial-aware Demand-guided Recursive Meshing for RGB-T Semantic Segmentation (Updating)

Tracker_SiamFC_DaSiamRPN

Object tracking based on SiamFC & DaSiamRPN using GOT-10k toolkit. Demo & Visualization.

ROS_Waypoints_Processor

Waypoints processor in pure ROS (Melodic)

Tools_Bag2PCD_ROS

Read the .bag files and save Lidar point clouds in .pcd files. Coordinate transformation is optional.

KITTI-Objects-Detection-Using-SSD-in-Tensorflow

This project was used to prepare a competition in the summer of 2018 ; The origin training structure is based on the SSD-Tensorflow by balancap.

Repository_Intelligent_Vehicle

This is a repository for the basic version of intelligent vehicles, which is on the basis of the trajectory tracking. It is mainly used to organize my open-sourced projects, and contains several related projects, researches, and tools of mine.

leofansq

ROS_OBU_DCU_Communication

This repository is for the communication between OBU & DCU, which is a simple V2X application.

AU326-Automatic-Puzzle-Solver

SJTU AU326: Automatic-Puzzle-Solver; 基于图像处理实现的自动拼图

Lane_Detection

Lane detection based on traditional method.

ROS_Object_Cluster

Object cluster in pure ROS (Melodic)

3D-Detection

Forked from AVOD & Add some new feature

DAIR-V2X

EI229-Image-BinaryClassification

SJTU EI229: Image Classification based on machine learning; 基于CART和C4.5的图像二分类

leofansq.github.io