Graph-Based Authentication Design for Color-Depth-Based 3D Video Transmission over Wireless Networks

3D video applications such as 3D-TV and 3D games have become more and more popular in recent years. These applications raised significant challenges in the media security, processing and transmissions. Especially, when 3D videos are delivered over wireless networks, the video streaming suffers the potential malicious attacks. One of the most important security challenging issues is how to guarantee the integrity of media content over error-prone wireless networks. To address this challenge, in the paper, we for the first time propose an authentication approach for 3D video transmission over wireless networks, which can improve the reconstructed media quality under error-prone wireless environment with lower authentication overheads and energy consumption. The proposed method is based on color-depth 3D video coding approach, which can save bandwidth, be tolerable to packet losses and thus satisfy the users' Quality of Experience (QoE) requirements. Our major contribution in this paper includes: (1) designing a joint source-channel-authentication coding framework for color-depth-based 3D video transmission; (2) proposing a media quality prediction model for color-depth-based 3D video transmission; (3) developing optimization for graph-based authentication on 3D video transmission to improve reconstructed media quality, reduce authentication overheads and energy consumption. Experimental results demonstrated the effectiveness of our proposed solutions.