Protecting Semantic Segmentation Models by Using Block-wise Image Encryption with Secret Key from Unauthorized Access

TL;DR

This paper introduces a novel method for protecting semantic segmentation models using block-wise image encryption with a secret key, ensuring access control while maintaining model performance for authorized users.

Contribution

It is the first to apply block-wise image transformation with a secret key for semantic segmentation model protection, extending prior classification-focused methods.

Findings

Authorized users access full model capacity with correct key.

Protected models show slight performance decline.

Unauthorized access results in significant performance deterioration.

Abstract

Since production-level trained deep neural networks (DNNs) are of a great business value, protecting such DNN models against copyright infringement and unauthorized access is in a rising demand. However, conventional model protection methods focused only the image classification task, and these protection methods were never applied to semantic segmentation although it has an increasing number of applications. In this paper, we propose to protect semantic segmentation models from unauthorized access by utilizing block-wise transformation with a secret key for the first time. Protected models are trained by using transformed images. Experiment results show that the proposed protection method allows rightful users with the correct key to access the model to full capacity and deteriorate the performance for unauthorized users. However, protected models slightly drop the segmentation performance compared to non-protected models.