Back to results
US 8,411,753 B2Granted
Color space scalable video coding and decoding method and apparatus for the same
AssigneeSamsung Electronics Co., Ltd.
InventorsSang-chang Cha, Manu Mathew, Kyo-hyuk Lee
Life of this patent
Estimating the term…
Abstract
A color space scalable video coding and decoding method and an apparatus for the same are disclosed that can adjust color components or color depth according to the performance of a decoder side. The color space scalable video coding method includes generating transform coefficients by removing the temporal redundancy and spatial redundancy of input video frames, quantizing the transform coefficients, generating a bit stream by entropy coding the quantized transform coefficients, and generating a color space scalable bit stream that includes the bit stream and position information of luminance data in the bit stream.
Claims16
A color space scalable video coding method, the method comprising:
- (a) generating transform coefficients by removing temporal redundancy and spatial redundancy of input video frames;
- (b) quantizing the transform coefficients;
- (c) generating a bit stream by entropy coding the quantized transform coefficients, and
- (d) generating a color space scalable bit stream comprising the bit stream and position information of luminance data in the bit stream,
- wherein the position information of the luminance data is inserted into a header of quantized texture data, and
- wherein the position information of the luminance data comprises a length of texture information in the bit stream, and a length of the luminance data in the texture information,
- wherein the length of texture information in the bit stream corresponds to number of layers in a Fine Grain SNR Scalability (FGS) layer structure.
4 more elements
A color space scalable video coding method comprising the steps of:
- (a) generating transform coefficients by removing temporal redundancy and spatial redundancy of input video frames;
- (b) quantizing the transform coefficients;
- (d) generating a color space scalable bit stream which comprises the bit stream and position information of luminance data in the bit stream,
- (e) wherein the position information of the luminance data is inserted into a header of quantized texture data;
- (f) generating color depth scalability based on Fine Grain Signal-to-Noise Ratio scalability (FGS) layer structure, comprising a first FGS layer, a second FGS layer and third FGS layer, and
- wherein a number of bits of a color depth of the bit stream is increased in proportion to an increase of a level of a layer.
3 more elements
A color space scalable video encoder comprising:
- a temporal transform unit that removes the temporal redundancy of input video frames;
- a spatial transform unit that removes the spatial redundancy of the input video frames;
- a quantization unit that quantizes the transform coefficients generated by the temporal transform unit and the spatial transform unit;
- an entropy coding unit that performs entropy coding of the quantized transform coefficients;
- and
- a color space scalable bit stream generation unit that generates a color space scalable bit stream that comprises a bit stream generated by the entropy coding unit, and
- position information of luminance data in the bit stream,
- wherein the position information of the luminance data is inserted into a header of quantized texture data,
- wherein the position information of a luminance data comprises a length of texture information in the bit stream, and a length of the luminance data in the texture information, and wherein the length of texture information in the bit stream corresponds to number of layers in a Fine Grain SNR Scalability (FGS) layer structure.
6 more elements
A method of transferring data of a slice that contains a plurality of macroblocks, comprising:
- (a) inserting position information of luminance data of all the macroblocks contained in the slice into a bitstream and inserting luminance data of all the macroblocks contained in the slice into the bitstream;
- (b) alternately inserting chrominance data of all the macroblocks contained in the slice into the bitstream after the inserted position information of the luminance data and between the luminance data,
- wherein the position information of the luminance data is inserted into a header of quantized texture data, and
- wherein the position information of the luminance data comprises a length of texture information in the bit stream, and a length of the luminance data in the texture information,
- wherein the length of texture information in the bit stream corresponds to number of layers in a Fine Grain SNR Scalability (FGS) layer structure;
- (c) transferring the bitstream that comprises the inserted position information of the luminance data, the inserted luminance data and the inserted chrominance data,
- wherein the slice is included in the fine granule SNR scalability (FGS) layer, and
- wherein the plurality of macroblocks do not contain motion vector information.
5 more elements
A method of transferring data of a slice that contains a plurality of macroblocks, comprising:
- (a) inserting position information of luminance data of all the macroblocks contained in the slice into a bitstream and inserting luminance data of all the macroblocks contained in the slice into the bitstream;
- (b) inserting chrominance data of all the macroblocks contained in the slice into the bitstream after the inserted position information of the luminance data and the luminance data,
- wherein the position information of the luminance data is inserted into a header of quantized texture data, and
- wherein the position information of the luminance data comprises a length of texture information in the bit stream, and a length of the luminance data in the texture information,
- wherein the length of texture information in the bit stream corresponds to number of layers in a Fine Grain SNR Scalability (FGS) layer structure;
- (c) transferring the bitstream that comprises the inserted position information of the luminance data, the inserted luminance data and the inserted chrominance data, and interpreting position of the luminance data and position of the chrominance data in the slice based on the position information of the luminance data,
- wherein the slice is included in the fine granule SNR scalability (FGS) layer.
4 more elements
Dependencies are inferred from the claim text.