Bit-ViP: Leveraging Bit-planes to Preserve Visual Privacy in Images through Obfuscation

2026-06-28

AI Summary

Existing image obfuscation methods often fail to balance privacy preservation with model usability—either they are vulnerable to model inversion attacks or produce images unsuitable for training. This work proposes a novel end-to-end trainable obfuscation scheme based on bit-plane decomposition, which innovatively integrates the Lorenz chaotic system and differential privacy. By injecting irreversible noise, the method effectively masks visual content while preserving semantic information crucial for downstream recognition tasks. Experimental results on the UCF101 and HMDB51 datasets demonstrate that the proposed approach significantly outperforms existing techniques in terms of resistance to reconstruction, pixel frequency distribution, information entropy, and inter-pixel correlation, thereby achieving a unified trade-off between security and trainability.

Abstract

The unprecedented growth of computer vision applications, such as surveillance systems and social media, raises security and visual privacy concerns, especially when data is stored on cloud servers. Image obfuscation offers a way to preserve visual privacy while maintaining an adequate level of usability; thus, it has been a topic of great interest in recent years. However, prior obfuscation schemes are either vulnerable to malicious attacks, such as model inversion to reconstruct original images from obfuscated images, or generate non-trainable obfuscated images, making them unusable for achieving reasonable accuracy. This paper proposes a novel bit-plane-based image obfuscation scheme, {\em Bit-ViP}, to preserve visual privacy for image-based recognition tasks. The Bit-ViP scheme produces secure, usable images by incorporating an innovative end-to-end obfuscation function. While doing so, the obfuscated image would contain non-invertible noise (generated by Lorenz's chaotic system and differential privacy), making it hard for an adversary to reconstruct the original image. We conduct extensive experiments on two popular activity recognition datasets, namely UCF101 and HMDB51, to validate the effectiveness of Bit-ViP. In the face of attacks on reconstruction, pixel frequency, information entropy, and pixel inter-correlation, we present a rigorous security analysis demonstrating tangible improvements over existing schemes.

Problem

• visual privacy
• image obfuscation
• model inversion
• usable privacy
• security

Innovation

• bit-plane obfuscation
• visual privacy
• chaotic system
• differential privacy
• non-invertible noise