Frigate

A complete and local NVR designed for Home Assistant with AI object detection. Uses OpenCV and Tensorflow to perform realtime object detection locally for IP cameras. 🛠️ SET-UP - You can configure your cameras directly within the Frigate app, there is no need to manually edit the configuration files. - Your RTSP password (if needed) is the "default app password" that is displayed on Frigate's page in the app store (shown after install).

⚠️ As usual, Frigate will attempt to update its configuration automatically. It is still recommended to back up your current config and database before upgrading. This release includes several major new features and improvements: - Face Recognition: Frigate can now detect and recognize faces, applying sub-labels to known faces. - License Plate Recognition (LPR): Frigate can now detect and recognize license plates, saving recognized plates to tracked car objects. - Viewer Role: A new role that restricts access to system features and the configuration editor. - Improved Support For Object Detection Hardware: Including updates for Hailo8, ONNX & OpenVINO, and RockChip (RKNN). - Camera Improvements: Including the ability to dynamically enable/disable cameras and select multiple streams per camera. - UI Changes: Including a rewritten review timeline, object path plotting, and edge/point snapping in the Mask/Zone editor. Full release notes are available at https://github.com/blakeblackshear/frigate/releases

• Version0.16.0
• CategoryHome & Automation
• Source codePublic
• Developed byBlake Blackshear
• Submitted by~dibref-labter
• Compatible withumbrelOS 0.5 or later