streamware

📚 StreamWare Documentation

Welcome to StreamWare documentation. This guide covers all features for real-time video analysis with AI.

📋 Table of Contents

Getting Started

• Installation
• Quick Start

Core Features

• 🎬 Real-time DSL Streaming - Live motion visualization in browser
• ⚡ Performance Optimization - GPU acceleration, timing logs, benchmarks
• 🤖 LLM Integration - Vision models, async inference
• 🎯 Motion Analysis - DSL-based tracking, blob detection
• 🧾 Accounting Scanner - Document scanning (web UI + RTSP, OCR)

Architecture

• 🏗️ System Architecture - Multiprocessing, data flow
• 📡 API Reference - CLI options, configuration

Development

• 🔧 Refactoring Plan - Tracking optimization roadmap

Deployment

• 💾 USB/ISO Builder - Bootable offline LLM environments

---

🚀 Quick Reference

Basic Commands

# Real-time viewer with LLM (recommended)
sq live narrator --url "rtsp://..." --realtime --turbo --fps 5

# DSL-only mode (fastest, no LLM)
sq live narrator --url "rtsp://..." --dsl-only --realtime --fps 20

# Full analysis with TTS
sq live narrator --url "rtsp://..." --mode track --tts

Key Options

Option	Description
--realtime	Enable browser viewer at http://localhost:8766
--dsl-only	Skip LLM, use only OpenCV tracking
--fps N	Target frames per second
--turbo	Skip checks + fast model + aggressive caching
--verbose	Show detailed timing logs

Performance Modes

Mode	FPS	LLM	Use Case
--dsl-only --fps 20	20	❌	Fast motion detection
--realtime --fps 5	5	✅ (async)	Balanced analysis
--turbo	2-5	✅ (fast)	Quick setup

---

📁 File Structure

streamware/
├── docs/                      # 📚 Documentation
│   ├── README.md              # This file
│   ├── REALTIME_STREAMING.md  # Real-time viewer guide
│   ├── PERFORMANCE.md         # Performance optimization
│   ├── LLM_INTEGRATION.md     # LLM configuration
│   ├── MOTION_ANALYSIS.md     # Motion tracking
│   └── ARCHITECTURE.md        # System design
├── streamware/
│   ├── components/
│   │   └── live_narrator.py   # Main narrator component
│   ├── frame_diff_dsl.py      # DSL motion analysis
│   ├── dsl_streamer_process.py # Separate DSL process
│   ├── realtime_dsl_server.py # WebSocket server
│   ├── fast_capture.py        # RTSP capture
│   ├── dsl_timing_logger.py   # Performance logging
│   └── async_llm.py           # Async LLM inference
└── README.md                  # Project overview

---

🔗 Related Links

• GitHub Repository
• Main README

This site is open source. Improve this page.