Frigate
Process security camera feeds locally with real-time AI object detection. Reduce false alerts, set up custom zones, and keep your footage completely private.
Protect your home and maintain complete privacy with an open-source network video recorder designed for local processing. By utilizing a supported AI accelerator, this system analyzes your IP camera feeds directly on your own hardware, ensuring your video never leaves your network. It performs over 100 object detections per second, drastically reducing false positives caused by shadows, wind, or simple motion. Key benefits include: Local AI Processing: Keep your footage private and avoid cloud subscription fees. Advanced Object Detection: Accurately identify people, cars, and other objects of interest without spending hours fine-tuning basic motion sensors. Custom Alert Zones: Define specific areas, like your driveway or front steps, to trigger precise notifications only when it matters. Seamless Automation Integration: Connect directly with Home Assistant, OpenHab, NodeRed, and MQTT to build powerful, low-latency home automation routines.
Directory Structure
docker-compose.yml
services:
frigate:
container_name: frigate
privileged: true
restart: unless-stopped
image: ghcr.io/blakeblackshear/frigate:stable
shm_size: "64mb"
devices:
- /dev/bus/usb:/dev/bus/usb
- /dev/dri/renderD128
volumes:
- /etc/localtime:/etc/localtime:ro
- ./config:/config
- ./storage:/media/frigate
- type: tmpfs
target: /tmp/cache
tmpfs:
size: 1000000000
ports:
- "5000:5000"
- "8554:8554"
- "8555:8555/tcp"
- "8555:8555/udp".env
# No sensitive environment variables are required for the default configuration.
# Frigate configuration is primarily handled via the config.yml file.Tags:
Similar to Frigate