초록 ▼

A method for encoding and decoding a video signal having frames with blocks comprised of pixels. The method reduces the complexity of the video signal to be encoded by applying a filter scheme to the input data stream in an encoder and later reversing the filter scheme in a decoder. A number of filt

A method for encoding and decoding a video signal having frames with blocks comprised of pixels. The method reduces the complexity of the video signal to be encoded by applying a filter scheme to the input data stream in an encoder and later reversing the filter scheme in a decoder. A number of filter schemes may be tested to determine the best filter scheme that most reduces the number of distinct color values per color plane. The best filter scheme is used to filter the data to be encoded. Filter information of the best filter scheme is included in the encoded data to enable the decoder to identify the best filter scheme in order to apply the best filter scheme in reverse to re-create the original data stream.

대표청구항 ▼

1. A method for encoding video data, comprising: applying a plurality of filter schemes to at least a portion of an input video signal to generate a plurality of filter outputs, each filter output having data indicating color values included in the filter output;determining, by a processor, the colo

1. A method for encoding video data, comprising: applying a plurality of filter schemes to at least a portion of an input video signal to generate a plurality of filter outputs, each filter output having data indicating color values included in the filter output;determining, by a processor, the color variance of at least some of the plurality of filter outputs;selecting at least one of the plurality of filter schemes based on the color variance of the filter output generated using the selected filter scheme;encoding the filter output corresponding to the at least one selected filter scheme;wherein encoding the filter output corresponding to the at least one selected filter scheme includes:generating an encoded signal, the encoded signal including information identifying the at least one selected filter scheme. 2. The method of claim 1, wherein determining the color variance of a filter output further comprises: counting the number of different color values in the filter output. 3. The method of claim 1, wherein determining the color variance of a filter outputs further comprises: determining the statistical dispersion of color values in at least a portion of the filter output. 4. The method of claim 1, wherein determining the color variance of a filter outputs further comprises: determining if the color values in at least a portion of the filter output include a dominant color value. 5. The method of claim 1, wherein the input video signal includes an image that has been anti-aliased using a blurring filter. 6. The method of claim 1, wherein the input video signal includes images of text that have been anti-aliased using a blurring filter. 7. The method of claim 1, wherein the plurality of filter schemes includes at least one edge-sharpening filter. 8. The method of claim 1, wherein the plurality of filter schemes includes convolution kernels having differing numbers of taps. 9. The method of claim 8, wherein the plurality of filter schemes includes filter schemes having 3, 5, 7 and 9 taps. 10. The method of claim 8, wherein the plurality of filter schemes include filter schemes having different weights. 11. The method of claim 1, wherein the portion of the input video signal is a block. 12. The method of claim 1, wherein the portion of the input video signal is a frame. 13. The method of claim 1, wherein the portion of the input video signal is selected based on the sharpness of the image represented by the input video signal. 14. The method of claim 1, wherein applying a plurality of filter schemes to the portion of the input video signal includes applying the plurality of filter schemes to a color plane of the portion of the input video signal. 15. A method for encoding data representing an image that is intended for display on a computer monitor, comprising: applying a plurality of filter schemes to at least a portion of an input video signal to generate a plurality of filter outputs;identifying, by a processor, a filter output from the plurality of filter outputs that has a low color variance relative to the other filter outputs from the plurality of filter outputs;generating information identifying the filter scheme used to generate the selected filter output;wherein a filter output from the plurality of filter outputs includes a plurality of values corresponding to a plurality of pixels; andencoding the identified filter output. 16. The method of claim 15, wherein: the input video signal includes a plurality of blocks representing text in the image that has been anti-aliased using a blurring filter; andthe portion of the input video signal includes at least some of the plurality of blocks. 17. The method of claim 15, wherein the plurality of filter schemes includes convolution kernels having differing numbers of taps. 18. A method for reducing the complexity of a data stream to be encoded, comprising: identifying at least one target block from an input video stream;obtaining an original set of color values that contains each distinct color value used in the at least one target block;applying, by a processor, a plurality of filter schemes to the at least one target block to generate a plurality of filter outputs;determining a filtered set of color values for each of the plurality of filter outputs, each filtered set of color values containing each distinct color value used in its associated filter output;selecting the filter scheme of the filter output that most reduces the complexity of the at least one target block; andgenerating compressed bitstream by encoding the filter output associated with the selected filter scheme. 19. The method of claim 18, further comprising: determining a difference between the number of color values in the original set of color values and each of the filtered set of color values; andidentifying the filter output that most reduces the complexity of the at least one target block as the filter output that is associated with the filtered set of color values that has the largest difference. 20. The method of claim 18, wherein encoding the filter output associated with the selected filter scheme includes: encoding a filter information identifier into the compressed bitstream identifying which filter scheme was used to create the filter output in the compressed bitstream.