A VVC Video Steganography Based on Coding Units in Chroma Components with a Deep Learning Network

Versatile Video Coding (VVC) is the latest video coding standard, but currently, most steganographic algorithms are based on High-Efficiency (HEVC). The concept of symmetry often adopted in deep neural networks. With rapid rise new multimedia, steganography shows great research potential. This paper proposes a VVC algorithm Units (CUs). Considering novel techniques VVC, proposed only uses chroma CUs to embed secret information. Based modifying partition modes CUs, we propose four different embedding levels satisfy needs visual quality, capacity and bitrate. In order reduce bitrate stego-videos improve distortion caused by them, convolutional network (CNN) as an additional in-loop filter codec achieve better restoration. Furthermore, components has advantage resisting steganalysis algorithms, since few have been thus far HEVC luminance component. Experimental results show that achieves excellent performance cost capacity.