👁️ Detect and log potential shoplifting events with Aisle Guard, a lightweight computer vision system for analyzing retail camera footage.

👥 Detect and count humans in videos or images using OpenCV with simple commands for effective monitoring and analysis.

Ultra-lightweight human detection. The number of parameters does not correlate to inference speed. For limited use cases, an input image resolution of 64x64 is sufficient. High-level object detection architectures such as YOLO are overkill.

Python+YOLOv8-based human/animal/object detection DVR framework with GUI, webUI and Telegram alerts

High-performance surveillance video human detector. Auto-detect people in videos, extract & merge clips. Multi-process pipeline with FFmpeg multi-threaded decoding. Supports Xiaomi, Hikvision, Dahua cameras. 高性能监控视频人形检测工具。自动检测视频中的人物，提取并合并片段。

Real-time human counting using YOLOv8

Aisle Guard is a lightweight, end-to-end computer vision system for detecting and logging potential shoplifting events from retail camera footage.

FastestDetNext: Various improvements have been made to make FastestDet even lighter and faster.

YOLOv8 trained on CrowdHuman dataset and supports ONNX Runtime Inference for Image, Video and Webcam

Real-time WiFi-based human presence detection with multi-person capability

A Human Detection and Tracking System with RE-ID using Yolov12n

Ultra-lightweight human detection. The number of parameters does not correlate to inference speed. For limited use cases, an input image resolution of 64x64 is sufficient. High-level object detection architectures such as YOLO are overkill.

This figure illustrates the advanced, five-stage data processing pipeline of the Advanced Falling Object Detection System (AFODS), as described in the accompanying manuscript. The flowchart details the sequential process designed for proactive threat detection, moving from initial sensor data acquisition to the final decision and action stage.

Flutter-based mobile IP camera that streams Android camera frames to a Python YOLOv8 server for real-time human detection, with live bounding-box overlay on both phone and web.

ORCUS - Multi-drone kamikaze system for autonomous area surveillance and target engagement. Intelligent grid partitioning prevents collisions during parallel operations. Drones scan zones, detect human targets with AI vision, lock on with precision tracking, and execute collision. ArduPilot SITL + ROS + Gazebo simulation.

A computer vision tool for detecting humans in images and video streams. Built with Python and OpenCV, it provides a simple and efficient way to perform people counting and surveillance

This repository contains a code for human detection and congestion detection using OpenCV and an Arduino UNO

Automatic fall detection system using YOLOv8 and MediaPipe. Detects people in videos, analyzes body posture, and automatically clips falling moments from the original video.

An Offline AI powered android app which can detect any human using the phone camera and send a SMS alert to the owner.