Securing Immersive 360 Video Streams through Attribute-Based Selective Encryption

2025-05-07

AI Summary

To address the high computational overhead, low cache efficiency, and poor scalability of secure high-bitrate interactive 360° video streaming over HTTPS, this paper proposes a viewport-adaptive hierarchical selective encryption framework. The framework integrates attribute-based encryption (ABE) with tile-level selective encryption: leveraging viewport prediction, it dynamically identifies visually critical tiles and assigns differentiated encryption strengths based on their relative field-of-view contribution—applying strong encryption to central tiles and lightweight encryption or plaintext transmission to peripheral ones. Evaluated on the CloudLab platform, the framework significantly reduces intermediate-node computational load compared to HTTPS, improves cache hit rates, and preserves VMAF quality without degradation. It simultaneously achieves strong security guarantees, real-time performance, and system scalability.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Hardware-aware ML
• Search and Optimization: Distributed Search

Application Category

• Security and Privacy: Large-scale security measurements
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Delivering high-quality, secure 360{deg} video content introduces unique challenges, primarily due to the high bitrates and interactive demands of immersive media. Traditional HTTPS-based methods, although widely used, face limitations in computational efficiency and scalability when securing these high-resolution streams. To address these issues, this paper proposes a novel framework integrating Attribute-Based Encryption (ABE) with selective encryption techniques tailored specifically for tiled 360{deg} video streaming. Our approach employs selective encryption of frames at varying levels to reduce computational overhead while ensuring robust protection against unauthorized access. Moreover, we explore viewport-adaptive encryption, dynamically encrypting more frames within tiles occupying larger portions of the viewer's field of view. This targeted method significantly enhances security in critical viewing areas without unnecessary overhead in peripheral regions. We deploy and evaluate our proposed approach using the CloudLab testbed, comparing its performance against traditional HTTPS streaming. Experimental results demonstrate that our ABE-based model achieves reduced computational load on intermediate caches, improves cache hit rates, and maintains comparable visual quality to HTTPS, as assessed by Video Multimethod Assessment Fusion (VMAF).

Problem

• Securing high-bitrate 360° video streams efficiently
• Reducing computational overhead in immersive media encryption
• Enhancing viewport-adaptive security without quality loss

Innovation

• Attribute-Based Encryption for 360 video
• Selective encryption reduces computational overhead
• Viewport-adaptive encryption enhances security