Digital watermarking has been proposed as very useful technology in the protection of digital data such as image, audio, video, formatted documents (PDF or PS), and 3D objects. In the literature, most of the existing watermarking approaches are conducted on images. However, video is even more useful and should be protected with higher priority. In particular, video sequences usually contain rich properties that images do not have. On the other hand, the types of attacks applied on a video are much different from those applied on an image. In this chapter, we will focus ourselves specifically on video watermarking. The content of this chapter is divided into two parts. In the first part, the existing video watermarking techniques are briefly reviewed. We have pointed out their advantages and disadvantages to realize what can be done about video watermarking. In the second part, we propose a compressed domain video watermarking scheme for copyright protection. For the sake of real-time video watermark detection, our method is directly conducted in the MPEG-2 bitstream. More specifically, watermarks are inserted into the VLC domain. We shall discuss how to select proper data in a video bitstream to embed watermarks while preserving perceptual fidelity. In addition, video watermarks are embedded by a new proposed watermarking technique, which is based on the concept of communications with side information. The power of our method is reflected by its robust capability against attacks. Future work will also be pointed out to further improve the current scheme
