VizMux
Lightweight, reliable motion detection for your IP cameras. No video recording, no heavy NVR — just fast snapshots when it matters.
Connects to any RTSP camera, runs server-side motion detection with configurable sensitivity and polygon masks, and outputs snapshot images on motion events. Pairs with PiyoAI for AI-powered object detection.
*Starting price. See full pricing & tiers.
**All security camera footage shown in the demo is mock data.
Everything You Need for Camera Motion Detection
Purpose-built to be lightweight and reliable. VizMux focuses on what it does best and delegates AI inference to PiyoAI.
Multi-Camera RTSP
Connect unlimited IP cameras via RTSP. Substream-first design keeps bandwidth and CPU usage low.
3 Detection Methods
EMA, MOG2, or KNN — each optimized for different environments. Selectable per camera with data-driven recommendations.
Polygon Masks
Draw exclusion zones directly in the browser. Normalized coordinates survive resolution changes.
Camera Scheduling
Set time windows per camera. Only detect motion when you need it — save resources and reduce noise.
Blue Iris Import
Already have a Blue Iris NVR? Import your cameras with one click — RTSP URLs auto-detected.
PiyoAI Integration
Pair with PiyoAI for YOLO object detection, MQTT alerts, and Telegram notifications. PiyoAI can run on a separate machine on your network.
Self-Update
One-click updates from inside the app using your download token. Progress tracking and automatic service restart.
IP Allowlist
Lock down access to trusted IPs only. Built-in ASGI middleware — no external firewall needed.
Stats & Analytics
24-hour charts, per-camera time series, automated tuning tips, and detection method recommendations.
See It in Action
A modern, dark-themed web UI that runs in any browser. Manage all your cameras from one place.
Your Cameras at a Glance
A real-time grid of all connected cameras with live thumbnails, colored status badges (running, motion detected, stopped, error), FPS counters, and one-click snapshot capture.
- Persistent thumbnails with one-click capture
- Real-time status: running, motion, stopped, error
- FPS and capture backend displayed per camera
Fine-Tuned Per Camera
Configure every aspect of each camera: RTSP URL, capture backend (FFmpeg or OpenCV), rotation, ONVIF stream probing, and detailed motion detection settings.
- Interactive polygon mask editor (draw exclusion zones)
- Sensitivity presets: Low, Medium, High, Very High
- ONVIF auto-probe for stream profile discovery
Browse Every Motion Event
A two-panel event browser with server-side pagination and camera filtering. Click any event to see the snapshot with motion bounding boxes overlaid.
- Scrollable list with timestamp, score, and thumbnail
- Full-size snapshot with fullscreen zoom
- Metadata: motion score, box count, active masks
Data-Driven Tuning
Understand your motion patterns with 24-hour charts, per-camera time series, and automated tuning tips that help you pick the right detection method.
- Pie chart of events by camera (click to drill down)
- Per-camera events/hour and motion score trends
- Automated EMA vs MOG2 vs KNN recommendations
Quick Setup, Deep Control
Tabbed settings for server, output paths, Blue Iris integration, PiyoAI connection, and self-updates. Import cameras from Blue Iris in seconds.
- One-click Blue Iris camera import
- PiyoAI local/remote mode auto-detection
- In-app self-update with progress tracking
Watch Any Camera Live
Full-screen MJPEG live streaming with camera selector, FPS counter, and detection method badge. Check what’s happening right now without leaving the app.
- Full-width MJPEG stream in the browser
- Camera selector and status overlay
- No plugins needed — works in any modern browser
Correlation Analytics
When VizMux and PiyoAI run on the same machine, VizMux unlocks Correlation Analytics — a dedicated page that compares VizMux motion detections against PiyoAI inference results side by side.
- Per-camera positive rate: how many motion events had real objects
- Hourly time series comparing VizMux detections vs PiyoAI positives
- Object class breakdown per camera per hour
- Tune motion sensitivity based on actual detection accuracy
A 33% positive rate means 67% of motion events were false alarms — time to raise thresholds or add masks.
Take a Tour
Watch VizMux in action — from the camera dashboard through events, stats, and correlation analytics.
How It Works
VizMux handles motion detection. PiyoAI handles AI. A clean, reliable pipeline.
1. RTSP Ingest
Cameras stream low-bandwidth substreams to VizMux via RTSP
2. Motion Detection
EMA, MOG2, or KNN analyzes frames with masks and lightning guard
3. Snapshot Output
Motion triggers a JPEG snapshot + JSON sidecar to disk
4. PiyoAI (YOLO)
PiyoAI watches the output folder and runs AI object detection
Built for Performance
Detection Methods Compared
| Method | Best For | CPU |
|---|---|---|
| EMA | Indoor / stable lighting | Lightest |
| MOG2 | Outdoor / changing light (sunrise, sunset) | Moderate |
| KNN | Dynamic scenes (water, flags, foliage) | Moderate |
Ready to Get Started?
VizMux starts at $20.20*. Pair it with PiyoAI for the full AI-powered camera monitoring stack from $40.20*.
*Basic tier. See all tiers on the pricing page.