This directory contains interactive desktop applications using OpenVINO as an engine for the AI inference.
The app blurs the background behind a person on a webcam, and it's built using NodeJS and Electron technologies. It can be run from the compiled exe file or using npm.
| Hide Your Mess Behind | |
|---|---|
| Related AI concepts | image segmentation |
| Platforms | Client (CPU, GPU, NPU), Edge (CPU) |
| Owner | @Roszczyk, @adrianboguszewski |
Control your computer with hand gestures using real-time hand detection. Features multiple application modes including presentation control, media playback, gaming, and browser navigation.
| PalmPilot - Gesture Control | |
|---|---|
| Related AI concepts | hand detection, gesture recognition, computer vision |
| Platforms | Client (CPU, GPU), Edge (CPU) |
| Owner | @vijaykr338 |
The Smart Intersection Traffic Monitoring System is an advanced AI-powered desktop application built with PySide6 and OpenVINO for real-time traffic monitoring, violation detection, and smart intersection analytics. The system combines computer vision, machine learning, and IoT technologies to provide comprehensive traffic management capabilities with enterprise-grade features.
| Traffic Monitoring System | |
|---|---|
| Related AI concepts | object detection, VLM |
| Platforms | Client (CPU, GPU, NPU), Edge (CPU) |
| Owner | @JatinSachdeva2004 |
This Telegram RAG Assistant is a privacy-first desktop app that locally processes Telegram data using OpenVINO-accelerated retrieval-augmented generation and a native Qt interface. It supports streaming responses, device diagnostics, and metadata-preserving chunking for efficient, secure AI-powered insights.
| Messenger AI‑Assistant | |
|---|---|
| Related AI concepts | LLM, RAG, embedddings, reranking |
| Platforms | Client (CPU, GPU, NPU) |
| Owner | @siddhant-0707 |
VisionGuard is a privacy-focused desktop app that uses your webcam and computer vision to track screen time and reduce eye strain with smart break reminders. Built for performance and efficiency, it runs locally on CPU, GPU, or NPU without needing internet access.
| VisionGuard | |
|---|---|
| Related AI concepts | gaze estimation |
| Platforms | Client (CPU, GPU, NPU), Edge (CPU) |
| Owner | @inbasperu |
Estimates crowd density from images or video streams using a CSRNet model converted to OpenVINO IR format — no ONNX intermediate step required.
| Crowd Density Estimation | |
|---|---|
| Related AI concepts | crowd counting, density estimation, computer vision |
| Platforms | Client (CPU, GPU), Edge (CPU) |
| Owner | @arunimakanavu |
