# De-identification without losing faces

## TL;DR

This paper introduces a face de-identification method that preserves facial attributes like expressions while concealing identities, using attribute transfer models and a small set of donor faces to ensure natural appearance and privacy protection.

## Contribution

The proposed method uniquely combines facial attribute transfer with limited donor faces to achieve high-quality de-identification without losing essential facial features.

## Key findings

- Effective privacy protection in images and videos.
- High visual quality of de-identified faces.
- Preservation of facial attributes like expressions.

## Abstract

Training of deep learning models for computer vision requires large image or video datasets from real world. Often, in collecting such datasets, we need to protect the privacy of the people captured in the images or videos, while still preserve the useful attributes such as facial expressions. In this work, we describe a new face de-identification method that can preserve essential facial attributes in the faces while concealing the identities. Our method takes advantage of the recent advances in face attribute transfer models, while maintaining a high visual quality. Instead of changing factors of the original faces or synthesizing faces completely, our method use a trained facial attribute transfer model to map non-identity related facial attributes to the face of donors, who are a small number (usually 2 to 3) of consented subjects. Using the donors' faces ensures that the natural appearance of the synthesized faces, while ensuring the identity of the synthesized faces are changed. On the other hand, the FATM blends the donors' facial attributes to those of the original faces to diversify the appearance of the synthesized faces. Experimental results on several sets of images and videos demonstrate the effectiveness of our face de-ID algorithm.