object-tracking

JavaScript object detection lightweight library for augmented reality (WebXR demos included). It uses convolutional neural networks running on the GPU with WebGL.

Object detection, tracking, and 6DoF pose estimation in the web browser, integrated training environment to train your own neural network models 🚀

Lightweight WebGL and JavaScript library for real time object detection, tracking and 6DoF pose estimation in the browser 📦. Included in WebAR.rocks.train for custom neural network training.

AI-powered visual reasoning tools for broadcast & ProAV. PTZ camera tracking, object detection, scene analysis using Moondream VLM. By StreamGeeks & PTZOptics.

Auto Object Documentation - JavaScript

A web app for multi-object tracking using YOLO and ByteTrack with ONNX Runtime Web

Observe object changes deeply without cloning in the most efficient manner.

An easy lightweight vanilla JavaScript SVG manager, fill/stroke manipulator, and external loader with caching.

Computer-vision pipeline for pickleball footage: object detection, tracking, and analysis, split into modular vision, ML, API, and frontend layers.

Traffic monitoring from webcam

Real-time browser-based computer vision dashboard with COCO-SSD object detection, lightweight tracking, performance metrics, and simulated robot telemetry.

VisionForge is a desktop-style computer vision workspace for custom YOLO workflows and more pretrained models. This project was developed as part of LEVEL UP AI | ROKETSAN Artificial Intelligence Hackathon - AI Workshop Day 2, under the From Image to Decision – Intelligent Object Detection and Tracking System with YOLO challenge.

The first phase, involving comprehensive analysis and visualization, has been successfully completed, yielding critical insights into traffic patterns and accident hotspots. The current phase involves developing an advanced Computer Vision model that utilizes CCTV data to identify and analyze misuse and irregularities in traffic behavior.

AI-powered littering detection and enforcement system: YOLOv8 + SORT tracking spots litter, FaceNet face recognition and TrOCR license-plate reading identify offenders, and a local LLM drafts incident reports. FastAPI/PostgreSQL/AWS S3 backend with a React dashboard

A web-implemented real-time object recognition application, using Tensorflow.js and React.js. The object detection is done via single-shot detection(ssd). Our method required shorter time which is a priority for real time recognition, compared to different other methods, through the whole process of capturing, detecting and recognising objects.

Real time parking occupancy for KSU with YOLOv8 + ByteTrack, FastAPI and React. Regional winner, national finalist at the Bank of America Code-a-Thon 2026.

AI-powered smart surveillance system for real-time scene understanding, person tracking, action recognition, intelligent alerts, and natural-language investigation.

YOLO-based parking booth occupancy tracking with per-vehicle dwell time and arrival counting.

Multi-camera ArUco marker tracking and QR-code timecode synchronization pipeline for learning-by-demonstration capture. Estimates 6DoF pose, exports synchronized CSV datasets, and visualizes 3D trajectories. Code used for TOFAS tracking.