DiffSLVA: Harnessing Diffusion Models for Sign Language Video Anonymization

TL;DR

This paper introduces DiffSLVA, a novel zero-shot, diffusion model-based method for sign language video anonymization that preserves linguistic content without relying on pose estimation, enabling real-world application.

Contribution

We develop DiffSLVA, leveraging diffusion models and ControlNet with edge detection and facial expression modules for effective sign language video anonymization in the wild.

Findings

Effective signer anonymization demonstrated in experiments

Preserves critical linguistic features of sign language videos

Operates without pose estimation, suitable for diverse datasets

Abstract

Since American Sign Language (ASL) has no standard written form, Deaf signers frequently share videos in order to communicate in their native language. However, since both hands and face convey critical linguistic information in signed languages, sign language videos cannot preserve signer privacy. While signers have expressed interest, for a variety of applications, in sign language video anonymization that would effectively preserve linguistic content, attempts to develop such technology have had limited success, given the complexity of hand movements and facial expressions. Existing approaches rely predominantly on precise pose estimations of the signer in video footage and often require sign language video datasets for training. These requirements prevent them from processing videos 'in the wild,' in part because of the limited diversity present in current sign language video datasets. To address these limitations, our research introduces DiffSLVA, a novel methodology that utilizes pre-trained large-scale diffusion models for zero-shot text-guided sign language video anonymization. We incorporate ControlNet, which leverages low-level image features such as HED (Holistically-Nested Edge Detection) edges, to circumvent the need for pose estimation. Additionally, we develop a specialized module dedicated to capturing facial expressions, which are critical for conveying essential linguistic information in signed languages. We then combine the above methods to achieve anonymization that better preserves the essential linguistic content of the original signer. This innovative methodology makes possible, for the first time, sign language video anonymization that could be used for real-world applications, which would offer significant benefits to the Deaf and Hard-of-Hearing communities. We demonstrate the effectiveness of our approach with a series of signer anonymization experiments.