Context-Aware Full Body Anonymization using Text-to-Image Diffusion Models
This addresses privacy protection for individuals in real-world datasets, such as those used in self-driving cars, by replacing full bodies to reduce recognition risks, though it is incremental as it builds on existing text-to-image diffusion models.
The paper tackles the problem of full body person anonymization in datasets by proposing a workflow that uses Stable Diffusion as a generative backend, outperforming state-of-the-art methods in image quality, resolution, Inception Score, and Frechet Inception Distance.
Anonymization plays a key role in protecting sensible information of individuals in real world datasets. Self-driving cars for example need high resolution facial features to track people and their viewing direction to predict future behaviour and react accordingly. In order to protect people's privacy whilst keeping important features in the dataset, it is important to replace the full body of a person with a highly detailed anonymized one. In contrast to doing face anonymization, full body replacement decreases the ability of recognizing people by their hairstyle or clothes. In this paper, we propose a workflow for full body person anonymization utilizing Stable Diffusion as a generative backend. Text-to-image diffusion models, like Stable Diffusion, OpenAI's DALL-E or Midjourney, have become very popular in recent time, being able to create photorealistic images from a single text prompt. We show that our method outperforms state-of-the art anonymization pipelines with respect to image quality, resolution, Inception Score (IS) and Frechet Inception Distance (FID). Additionally, our method is invariant with respect to the image generator and thus able to be used with the latest models available.