Now You See Me, Now You Don't: A Unified Framework for Expression Consistent Anonymization in Talking Head Videos

2026-01-14

AI Summary

This work proposes Anon-NET, a unified framework designed to preserve essential attributes—such as facial expression, head pose, age, gender, and ethnicity—for downstream visual tasks while effectively anonymizing identity in face videos. Anon-NET is the first approach to integrate diffusion generative models with video-driven facial animation, leveraging high-level attribute recognition and motion-aware expression transfer to achieve strong identity obfuscation without compromising expression consistency, visual realism, or temporal coherence. Extensive experiments on VoxCeleb2, CelebV-HQ, and HDTF datasets demonstrate that Anon-NET significantly outperforms existing methods in both visual quality and temporal stability while successfully concealing identity information.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation
• Machine Learning: Privacy

Application Category

• Security and Privacy: Data transparency and provenance
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Face video anonymization is aimed at privacy preservation while allowing for the analysis of videos in a number of computer vision downstream tasks such as expression recognition, people tracking, and action recognition. We propose here a novel unified framework referred to as Anon-NET, streamlined to de-identify facial videos, while preserving age, gender, race, pose, and expression of the original video. Specifically, we inpaint faces by a diffusion-based generative model guided by high-level attribute recognition and motion-aware expression transfer. We then animate deidentified faces by video-driven animation, which accepts the de-identified face and the original video as input. Extensive experiments on the datasets VoxCeleb2, CelebV-HQ, and HDTF, which include diverse facial dynamics, demonstrate the effectiveness of AnonNET in obfuscating identity while retaining visual realism and temporal consistency. The code of AnonNet will be publicly released.

Problem

• face video anonymization
• privacy preservation
• expression consistency
• talking head videos
• identity obfuscation

Innovation

• diffusion-based generative model
• expression-consistent anonymization
• video-driven animation
• motion-aware expression transfer
• facial attribute preservation