CFE-PPAR: Compression-friendly encryption for privacy-preserving action recognition leveraging video transformers

2026-05-07

AI Summary

Existing encrypted privacy-preserving action recognition methods suffer significant degradation in both performance and visual quality after video compression, making it challenging to balance privacy and practical utility. This work proposes CFE-PPAR, a compression-friendly encryption scheme that, for the first time, achieves compatibility with mainstream compression standards such as Motion-JPEG and H.264. By integrating key-driven video encryption with parameter-transformed video Transformers, CFE-PPAR enables direct action recognition without decrypting the original content. Experimental results demonstrate that CFE-PPAR substantially outperforms existing approaches on the UCF101 and HMDB51 benchmarks, simultaneously ensuring strong privacy guarantees, high recognition accuracy, and robust compatibility with video compression.

Abstract

Privacy-preserving action recognition (PPAR) enables machines to understand human activities in videos without revealing sensitive visual content. Among the various strategies for PPAR, encryption-based methods achieve strong privacy protection while maintaining high recognition performance. However, these methods lead to a catastrophic decrease in recognition performance and visual quality when the encrypted videos are compressed. That is, the previous methods are not compression-friendly. To address these issues, in this paper, we propose the first compression-friendly encryption method for PPAR, called CFE-PPAR. In CFE-PPAR, videos encrypted with secret keys can be directly recognized by a video transformer, which uses parameters transformed by the same keys as those used for video encryption. In experiments, it is verified that CFE-PPAR outperforms previous methods on the UCF101 and HMDB51 datasets under Motion-JPEG and H.264 compression.

Problem

• privacy-preserving action recognition
• compression-friendly
• video encryption
• video transformers
• compressed video

Innovation

• compression-friendly encryption
• privacy-preserving action recognition
• video transformers
• encrypted video recognition
• key-dependent parameter transformation