recops is an open-source facial analysis framework, an AI forensic toolkit designed specifically for visual investigations and analysis workflows in OSINT research. recops is developed by a duo of culture jamming activists based in Athens, and is intended for use by open source research agencies, journalists, and activists. recops leverages a whole range of complex operations from fundamental face detection and face recognition to face analysis and face clustering to bring practical applications to human rights research.
Through a combination of targeted data collection and collation, intelligent automation and analysis, the project establishes a new and comprehensive methodological framework to equip investigative journalists, human rights researchers, and digital activists with the means to strengthen justice and advocacy efforts.
recops is a wrapper around the open-sourced deepface library (a lightweight face recognition and facial attribute analysis library for Python) that stores data in a structured way using SQL. Wraps all cutting-edge artificial intelligence models for face detection and face recognition and automatically handles all procedures for facial recognition in the background.
recops is an open-source and in active development project, written in Python and licensed under the Do No Harm License. This means that anyone can use and adapt recops according to the challenges of their own research. It is lightweight and is mainly based on TensorFlow and Keras, that makes it very easy to install.
To learn more about the development context of the recops project, you can read an overview here.
The open-source recops software is developed by a duo of culture jamming activists, battling the opaque and intricate apparatus of our reality, based in Athens.
Email is the best way of contacting us. Please use our public PGP key to send us encrypted secure email using this address: recops@pm.me.
recops in its entirety was developed using open-source software. Makes heavy use of the deepface library - a lightweight face recognition and facial attribute analysis library for Python by Sefik Serengil.
Additionally makes use of sqlite as SQL database, sqlalchemy and its ORM capabilities, flask and flask-admin to build a helpfull Web UI.
recops is licensed under the Do No Harm License, see LICENSE.md for more details. Is a wrapper around deepface library. Also recops wraps some external face recognition models: VGG-Face, Facenet, OpenFace, DeepFace, Dlib, DeepID, ArcFace, and SFace.
Besides, age, gender and race / ethnicity models are based on VGG-Face. Licence types will be inherited if you are going to use those models. Please check the license types of those models for production purposes.
The site is built with Markdown, is designed to be privacy-friendly and does not use any 3rd party analytics to track visits, nor any 3rd party dependencies that compromise privacy or share data.