[论文解读] Convolutional Video Steganography with Temporal Residual Modeling

Xinyu Weng, Yongzhi Li+2 more|arXiv (Cornell University)|Jun 8, 2018

Advanced Steganography and Watermarking Techniques参考文献 19被引用 19

一句话总结

本文提出了一种新颖的深度卷积视频隐写框架，通过利用时间残差建模来挖掘帧间稀疏性，从而在不可察觉性和鲁棒性方面均取得提升。通过分别使用专用神经网络分支对参考帧和帧间残差进行编码，并结合基于阈值的路由机制，该模型在APD和隐写分析抵抗能力方面均优于LSB和图像隐写基线模型。

ABSTRACT

Steganography represents the art of unobtrusively concealing a secrete message within some cover data. The key scope of this work is about visual steganography techniques that hide a full-sized color image / video within another. A majority of existing works are devoted to the image case, where both secret and cover data are images. We empirically validate that image steganography model does not naturally extend to the video case (i.e., hiding a video into another video), mainly because it completely ignores the temporal redundancy within consecutive video frames. Our work proposes a novel solution to the problem of video steganography. The technical contributions are two-fold: first, the residual between two consecutive frames tends to zero at most pixels. Hiding such highly-sparse data is significantly easier than hiding the original frames. Motivated by this fact, we propose to explicitly consider inter-frame residuals rather than blindly applying image steganography model on every video frame. Specifically, our model contains two branches, one of which is specially designed for hiding inter-frame difference into a cover video frame and the other instead hides the original secret frame. A simple thresholding method determines which branch a secret video frame shall choose. When revealing the concealed secret video, two decoders are devised, revealing difference or frame respectively. Second, we develop the model based on deep convolutional neural networks, which is the first of its kind in the literature of video steganography. In experiments, comprehensive evaluations are conducted to compare our model with both classic least significant bit (LSB) method and pure image steganography models. All results strongly suggest that the proposed model enjoys advantages over previous methods. We also carefully investigate key factors in the success of our deep video steganography model.

研究动机与目标

• 为解决现有图像隐写模型在视频隐写中因忽略时间冗余性而存在的局限性。
• 利用帧间残差的稀疏性，实现更高效且不可察觉的视频数据隐藏。
• 开发一种基于深度学习的框架，联合优化低失真与高解码准确率。
• 通过引入对抗性训练与GAN风格判别器，增强对隐写分析的抵抗能力。
• 研究帧选择、阈值设定以及视频内容质量对隐写性能的影响。

提出的方法

• 模型采用两条并行分支：一条用于隐藏参考帧，另一条用于隐藏帧间残差，通过阈值机制将每帧路由至合适的分支。
• 训练一个参考帧或残差帧（RoR）分类器，以区分载体帧中包含的是参考数据还是残差数据，基于来自独立H/R-nets的概率。
• 框架采用GAN风格的对抗损失以提升对隐写分析的抵抗能力，通过超参数调节H/R损失的权重。
• 基于1000段视频剪辑的平均残差能量，经验性地将帧分割的阈值设定为30.68。
• 通过结合感知损失、重建损失和对抗损失进行模型训练，以最小化失真并提升检测抵抗能力。
• 将一个视频片段定义为一个参考帧及其后续的残差帧，分析片段长度分布以平衡模型性能。

实验结果

研究问题

• RQ1 与独立处理每一帧相比，能否利用视频中的帧间残差来提升隐写效率和不可察觉性？
• RQ2 将时间残差建模应用于视频时，其表现与标准图像隐写方法相比如何？
• RQ3 在模型性能与鲁棒性方面，区分参考帧与残差帧的最优阈值是多少？
• RQ4 使用GAN判别器进行对抗性训练在多大程度上提升了对隐写分析的抵抗能力？
• RQ5 哪些视频内容特征（如纹理、饱和度）与低APD得分及更优的隐写性能相关？

主要发现

• 经对抗性训练后，H/R-nets的APD得分分别达到4.13和6.21，表明对隐写分析的抵抗能力得到提升。
• 对抗性训练后的模型需超过600组训练样本才能达到80%的隐写分析准确率，而原始模型仅需400组，证明安全性显著增强。
• 当采用结合真实帧与伪造帧概率的基于概率的决策规则时，RoR分类器在测试集上达到100%准确率。
• 该模型在不可察觉性和解码保真度方面均优于经典的LSB和纯图像隐写模型。
• 低纹理与低饱和度的视频被一致识别为“优质”载体或秘密内容，与更低的APD得分相关。
• 30.68的帧分割阈值在载体质量与解码秘密保真度之间实现了平衡的权衡。