摘要 |
The present invention relates to a computer implemented method relating to a method for embedding depth information in a compressed color bit string, to disclose a method for extraction of embeddable information from a depth image based on a correlation between a watermark capacity and a color/depth image. The method comprises the following steps: (a) calculating a watermark capacity embeddable in a compressed color bit string by a predetermined information embedding technique; (b) determining the number of segmentations of a color image and the number of clusters to be used in clustering depth values of depth images having the same view point as that of the color image; (c) determining whether or not the number of bits of a watermark to be generated when applying the determined number of the segmentations and the determined number of the clusters exceeds the calculated watermark capacity; (d) generating a depth image reconfigured based on a clustering depth value applied to each image segment when the calculated watermark capacity is not exceeded based on a determination result; and (e) embedding, in the compressed color bit string, the bits of the watermark which is generated when applying the reconfigured depth image having the closest accuracy to an original depth image through quality comparison of the generated reconfigured depth image. |