Rendering-Refined Stable Diffusion for Privacy Compliant Synthetic Data
This addresses privacy regulations like GDPR for image datasets, particularly human-centric ones, by providing a pseudonymization method that retains utility, though it appears incremental as it builds on existing Stable Diffusion and 3D-rendering techniques.
The paper tackles the problem of privacy-compliant image pseudonymization by introducing Rendering-Refined Stable Diffusion (RefSD), which preserves posture and enables attribute control while maintaining realism. The result shows that models trained on RefSD pseudonymized data outperform those trained on real data in detection tasks, with further gains when combined with real data.
Growing privacy concerns and regulations like GDPR and CCPA necessitate pseudonymization techniques that protect identity in image datasets. However, retaining utility is also essential. Traditional methods like masking and blurring degrade quality and obscure critical context, especially in human-centric images. We introduce Rendering-Refined Stable Diffusion (RefSD), a pipeline that combines 3D-rendering with Stable Diffusion, enabling prompt-based control over human attributes while preserving posture. Unlike standard diffusion models that fail to retain posture or GANs that lack realism and flexible attribute control, RefSD balances posture preservation, realism, and customization. We also propose HumanGenAI, a framework for human perception and utility evaluation. Human perception assessments reveal attribute-specific strengths and weaknesses of RefSD. Our utility experiments show that models trained on RefSD pseudonymized data outperform those trained on real data in detection tasks, with further performance gains when combining RefSD with real data. For classification tasks, we consistently observe performance improvements when using RefSD data with real data, confirming the utility of our pseudonymized data.