Forbes: Face Obfuscation Rendering via Backpropagation Refinement Scheme

TL;DR

The paper introduces Forbes, a face obfuscation algorithm that makes faces unrecognizable to humans while remaining identifiable by machines, using a backpropagation-based optimization process.

Contribution

It presents a novel face obfuscation method that balances human privacy with machine recognition, combining multiple transformations with an optimization scheme.

Findings

Achieves high human indecipherability of faces.

Maintains machine recognizability of faces.

Demonstrates effectiveness across various datasets.

Abstract

A novel algorithm for face obfuscation, called Forbes, which aims to obfuscate facial appearance recognizable by humans but preserve the identity and attributes decipherable by machines, is proposed in this paper. Forbes first applies multiple obfuscating transformations with random parameters to an image to remove the identity information distinguishable by humans. Then, it optimizes the parameters to make the transformed image decipherable by machines based on the backpropagation refinement scheme. Finally, it renders an obfuscated image by applying the transformations with the optimized parameters. Experimental results on various datasets demonstrate that Forbes achieves both human indecipherability and machine decipherability excellently. The source codes are available at https://github.com/mcljtkim/Forbes.