nextcloud/recognize

👁 👂 Smart media tagging for Nextcloud: recognizes faces, objects, landscapes, music genres

68
/ 100
Established

Runs pre-trained deep learning models (EfficientNet v2, MoViNet, face-api.js, Musicnn) locally via TensorFlow.js in Node.js, processing media entirely on-premises without external cloud services. Integrates with Nextcloud's Collaborative Tags system to enable seamless browsing in the Memories photo app and Audioplayer app. Supports selective directory exclusion via dotfiles (.noimage, .nomusic, .novideo, .nomedia) and requires ~4GB RAM with native x86-64 or WebAssembly fallback for ARM processors.

676 stars. Actively maintained with 51 commits in the last 30 days.

No Package No Dependents
Maintenance 25 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

676

Forks

61

Language

PHP

License

AGPL-3.0

Last pushed

Mar 10, 2026

Commits (30d)

51

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/nextcloud/recognize"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.