| 摘要 |
A method, computer system, and a computer program product is disclosed for encoding a sequence of color frames. The method includes a processor obtaining a sequence of color frames and rendering a first frame of the sequence in at least three components, each component representing a different color value. For each of the at least three components, the processor performs a wavelet transform on the first frame, initializes a significant points list and a list of insignificant sets to create an individual spatial tree of the list of insignificant sets for each component, searches all of the at least three components to identify a most significant bit in the first frame, and creates a consolidated significance map with all the color components for the first frame by searching the individual spatial tree for each component. The processor then obtains this significance map and utilizes the significance map to encode the first frame. The processor retains the significance map in memory as a key frame significance map and retains the first frame, or a color transform of the first frame, in memory, as a key frame. |
| 主权项 |
1. A method for encoding a sequence of color frames, the method comprising:
obtaining, by a processor, a sequence of color frames; optionally creating a color transform for a first frame of the sequence of color frames; rendering, by a processor, in at least three components, a first frame of the sequence of color frames, wherein each component represents a different color value; for each of the at least three components, performing a wavelet transform on the first frame; for each of the at least three components, constructing a significant points list and a list of insignificant sets and creating an individual spatial tree of the list of insignificant sets for each component of the at least three components for the first frame; for each of the at least three components, initializing the significant points list and the list of insignificant sets; searching, by the processor, all of the at least three components to identify a most significant bit in the first frame; creating, by the processor, a significance map for the first frame by searching the individual spatial trees for each component of the at least three components, wherein the significance map represents all of the at least three color components; retaining, in a memory, the significance map as a key frame significance map, wherein the key frame significance map is utilized by the processor to encode a frame subsequent to the first frame in the sequence of color frames if the frame is within a given tolerance; and retaining, in the memory, one of: the first frame, or the color transform of the first frame, as a key frame. |
