CPC G06T 9/00 (2013.01) [G06V 10/751 (2022.01); G06V 10/762 (2022.01); G06V 10/764 (2022.01)] | 20 Claims |
1. A method comprising:
(a) receiving a source image, wherein the source image is one of: (i) an image file, (ii) a frame of a video;
(b) analyzing content of the source image, on a pixel-by-pixel basis; and classifying each pixel in the source image as either (I) a pixel associated with Photographic content, or (II) a pixel associated with Non-Photographic content;
(c) generating a pixel-clusters map that indicates (i) one or more clusters of pixels that were classified as associated with Photographic content of said source image, and (ii) one or more clusters of pixels that were classified as associated with Non-Photographic content of said source image;
(d) generating a composed image, by:
(d1) applying a first encoding technique to encode at least one of the clusters of pixels that were classified as associated with Photographic content of said source image;
(d2) applying a second, different, encoding technique to encode at least one of the clusters of pixels that were classified as associated with Non-Photographic content of said source image;
wherein the method is implemented by utilizing at least a hardware processor.
|