Privacy-Shielded Image Compression: Defending Against Exploitation from Vision-Language Pretrained Models

Xuelin Shen, Jiayin Xu, Kangsheng Yin, Wenhan Yang

Abstract: The improved semantic understanding of vision-language pretrained (VLP) models has made it increasingly difficult to protect publicly posted images from being exploited by search engines and other similar tools. In this context, this paper seeks to protect users' privacy by implementing defenses at the image compression stage to prevent exploitation. Specifically, we propose a flexible coding method, termed Privacy-Shielded Image Compression (PSIC), that can produce bitstreams with multiple decoding options. By default, the bitstream is decoded to preserve satisfactory perceptual quality while preventing interpretation by VLP models. Our method also retains the original image compression functionality. With a customizable input condition, the proposed scheme can reconstruct the image that preserves its full semantic information. A Conditional Latent Trigger Generation (CLTG) module is proposed to produce bias information based on customizable conditions to guide the decoding process into different reconstructed versions, and an Uncertainty-Aware Encryption-Oriented (UAEO) optimization function is designed to leverage the soft labels inferred from the target VLP model's uncertainty on the training data. This paper further incorporates an adaptive multi-objective optimization strategy to obtain improved encrypting performance and perceptual quality simultaneously within a unified training process. The proposed scheme is plug-and-play and can be seamlessly integrated into most existing Learned Image Compression (LIC) models. Extensive experiments across multiple downstream tasks have demonstrated the effectiveness of our design.

Lay Summary:

In today's digital landscape, images are widely shared and processed by AI systems capable of extracting detailed information. While this enables many valuable applications, it also raises significant privacy concerns, as individuals may unintentionally disclose sensitive content.
Our research proposes a novel image compression framework to tackle this challenge. The framework compresses images to preserve visual quality for human observers while making them less accessible to AI models, thereby enhancing privacy protection.
By embedding privacy considerations into the image compression pipeline, our approach provides a practical solution for individuals and organizations seeking to protect sensitive visual information in an increasingly AI-driven world.

Application-Driven Machine Learning: This submission is on Application-Driven Machine Learning.

Link To Code: https://github.com/JiayinXu5499/PSIC

Primary Area: Applications->Computer Vision

Keywords: Privacy Safeguard, Backdoor Attack, Vision-Language Pretrained Models, Learned Image Compression

Submission Number: 15467