Embedding data in and detecting embedded data from video objects
Methods for embedding data, such as digital watermarks, in video objects segmented from video frames. Watermarks are embedded such that they move with the object in the video frames. Interframe motion of video objects is estimated and used to modify the watermark accordingly. Methods for detecting embedded data from video objects are also disclosed.
1. A method of embedding data into video, the method comprising:
receiving video frames including video objects, wherein at least one of the video objects is segmented from the video frames;
embedding data into the at least one video object; and
combining the video objects, including the at least one video object with embedded data, into video frames to produce a video stream including the at least one video object separately embedded with data.
2. The method of claim 1 wherein interframe motion of the segmented video object is estimated and used to modify the embedded data in different video frames.
3. The method of claim 1 wherein the at least one video object has corresponding motion parameters that define the motion of the video object relative to at least a first frame.
4. The method of claim 3 wherein the video object is segmented into blocks and the blocks have corresponding motion parameters.
5. The method of claim 1 including computing distortion associated with re-inserting a watermark for a video object from a previous frame into the video object for a current frame, and embedding the watermark from the previous frame into the video object for the current frame when the distortion is below a predetermined threshold.
6. The method of claim 1 wherein unique watermarks are embedded in different video objects segmented from the video frames.
7. The method of claim 1 including estimating interframe motion.
8. A method of embedding a watermark into video, the method comprising;
receiving video frames, including at least one video object that has been segmented from the video frames;
receiving motion parameters that define motion of the video object in the video frames;
embedding a watermark in the video object in plural frames, including using the motion parameters so that location of the watermark in the video frames corresponds to motion of the video object.
9. The method of claim 8 including computing the watermark as function of a frequency domain transform of the video object in the frames.