Official code for "DiffLight: A Partial Rewards Conditioned Diffusion Model for Traffic Signal Control with Missing Data".

A V2G Simulation Environment for large scale EV charging optimization

A Reinforcement Learning-based Volt-VAR Control Dataset

Software to process DJI infrared (IR) pictures, with advanced thermal analysis, conversion and visualization tools. Support for DJI M30T, M3T and Mavic 2 Enterprise Advanced

DepthSplat: Connecting Gaussian Splatting and Depth

ThermoNeRF (Thermographic NeRF)

ECCV2024

一款基于深度学习，提供数据标注、模型训练、模型部署、基于已有模型自动标注等功能的计算机视觉一体化平台。 操作简单，无需深度学习理论即可实现AI落地。处理分类、目标检测、目标跟踪等计算机视觉任务。 应用场景：交通执法（车辆违停、行人乱穿马路等）、工业检测（缺陷检测、物体分拣等）、安防（越界、人员聚集、违禁品、拉横幅等）等领域。

Image-to-point cloud Registration Tool

The repo for "Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image" and "Metric3Dv2: A Versatile Monocular Geometric Foundation Model..."

[AAAI 2025] DepthFM: Fast Monocular Depth Estimation with Flow Matching

This repository has code to perform many tests on point cloud registration, optimization, rendering and color calibration using the Open3D library.

Pyside6 application for creating thermal point clouds using photogrammetry. Compatible with DJI M3T (Mavic 3 thermal) and Mavic 2 Enterprise Advanced. Using Agisoft Metashape Python API

Thermal Image to 3D Point Cloud Coregistration

A system to detect non-adopted persons using a deep learning model, identify faces, and record work logs

安全帽检测+跟踪

A robust registration method for UAV thermal infrared and visible images taken by the camera with two sensors

FusionGAN: A generative adversarial network for infrared and visible image fusion

可基于摄像头实时监控或录制的视频或静态图片进行行人检测(lffd)/跟踪(deep sort)和行人重识别(reid)。

A traditional way to solve pointer meter reading

an improved gauge reader, read digit result via its' photo

My Master's thesis source code and links to the annotated dataset

Use yolov5 to detect pointer scale

Using YoloV3 to detect pointer instruments and reading the number by Hough Transform

基于opencv的指针式仪表的识别与读数

Reinforcement Learning-based exploration algorithm for a fleet of UAVs in an unknown environment.

Common used path planning algorithms with animations.

PyTorch implementations of deep reinforcement learning algorithms and environments

深度强化学习路径规划, SAC-Auto路径规划, Soft Actor-Critic算法, SAC-pytorch，激光雷达Lidar避障，激光雷达仿真模拟，Adaptive-SAC

This is a new repo used for training UAV navigation (local path planning) policy using DRL methods.