← Directory

CompreFace

CompreFace is a free and open-source, Docker-deployed face recognition service that provides web APIs for face detection, recognition, verification, landmark,…

  • Home Automation
  • Productivity
  • Tools & Utilities
CompreFace is a free and open-source, Docker-deployed face recognition service that provides web APIs for face detection, recognition, verification, landmark, age, gender, mask, and head pose analysis for homelab users who want to integrate facial intelligence without machine-learning expertise.

Platform availability

Available on 1 of 9 platforms

  • Unraid
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)

Health score

Not enough data yet for a health score. It needs at least 2 of 4 factors: maintenance, popularity, ease of install, resource needs.

Availability: platform listings confirmed today, links checked 6 days ago

No GitHub repository data collected yet.

Resources & Compatibility

ARM (e.g. Raspberry Pi) is not supported.

Checked 14 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Install Docker and Docker Compose first Make sure Docker (the tool that runs isolated apps in 'containers') and Docker Compose (which starts several containers together) are installed before running any start script. If you skip this, the start command will fail and CompreFace won't come up. source
  • Start CompreFace with docker-compose up -d Run docker-compose up -d from the CompreFace folder to launch the app in the background. This is the main start command, so without it the services stay off and the web interface won't be reachable. source
  • Optional: faster face checks with a GPU If you have an NVIDIA GPU and want faster processing, you can build and run the CUDA/GPU image as described in the embedding calculator docs. If you don't have a compatible GPU, skip this and use the normal CPU setup. source
  • Model builds Choose a Docker image build: FaceNet (default), ArcFace-R100 (most accurate but slowest), or MobileNet (fastest CPU model). source
  • Database PostgreSQL is installed by default; an external PostgreSQL server is optional and requires a superuser account. source

Fetched 5 days ago - each note links to its source

Alternatives (1)

  • Pipl - Business solutions for decoding online identities while linking them to real-world records.

Checked 11 days ago - source: AlternativeTo

Common questions

How do I start CompreFace after setting up the Docker files?

Run docker-compose up -d from the CompreFace folder to launch the app in the background. If you skip this, the services stay off and the web interface is unreachable.

Can I use a GPU for faster face processing?

If you have an NVIDIA GPU, you can build and run the CUDA/GPU image for faster processing. If you do not have a compatible GPU, use the normal CPU setup.

Can I manage subjects and image collections in CompreFace 1.0 without writing scripts?

Yes, CompreFace 1.0 includes ways to manage subjects and image collections without scripts or code. You can also test REST endpoints with predefined methods.

What REST API features does CompreFace provide?

CompreFace provides a REST API for face recognition, face verification, face detection, landmark detection, age recognition, and gender recognition. It is described as integratable into other systems without prior machine learning skills.

Answers sourced from github.com, github.com, www.reddit.com, www.reddit.com

Community

Collected 13 days ago - public community threads