SLIC: Secure Learned Image Codec through Compressed Domain Watermarking to Defend Image Manipulation

TL;DR

SLIC introduces a neural network-based image codec that embeds watermarks in the compressed domain, enabling detection of tampering through quality degradation, thus enhancing image authenticity and security.

Contribution

It presents a novel active watermarking method in the compressed domain using neural networks, improving tamper detection and image integrity.

Findings

Effective watermark embedding in compressed images

Tampered images show detectable quality degradation

Balances invisibility with robustness in watermarking

Abstract

The digital image manipulation and advancements in Generative AI, such as Deepfake, has raised significant concerns regarding the authenticity of images shared on social media. Traditional image forensic techniques, while helpful, are often passive and insufficient against sophisticated tampering methods. This paper introduces the Secure Learned Image Codec (SLIC), a novel active approach to ensuring image authenticity through watermark embedding in the compressed domain. SLIC leverages neural network-based compression to embed watermarks as adversarial perturbations in the latent space, creating images that degrade in quality upon re-compression if tampered with. This degradation acts as a defense mechanism against unauthorized modifications. Our method involves fine-tuning a neural encoder/decoder to balance watermark invisibility with robustness, ensuring minimal quality loss for non-watermarked images. Experimental results demonstrate SLIC's effectiveness in generating visible artifacts in tampered images, thereby preventing their redistribution. This work represents a significant step toward developing secure image codecs that can be widely adopted to safeguard digital image integrity.