发明名称 Image coding method, image decoding method, image coding apparatus, image decoding apparatus, and image coding and decoding apparatus
摘要 An image coding method of coding an image on a per coding unit basis, the method comprising: applying a frequency transform to luminance data and chrominance data of transform units in the coding unit including predetermined blocks each corresponding to one or more of the transform units; and coding the luminance data and the chrominance data to which the frequency transform has been applied to generate a bitstream in which the luminance data and the chrominance data are grouped on a per predetermined block basis.
申请公布号 US9615099(B2) 申请公布日期 2017.04.04
申请号 US201514686838 申请日期 2015.04.15
申请人 SUN PATENT TRUST 发明人 Terada Kengo;Nishi Takahiro;Shibahara Youji;Tanikawa Kyoko;Sasai Hisao;Sugio Toshiyasu;Matsunobu Toru
分类号 G06K9/36;H04N19/136;G06T9/00;H04N19/70;H04N19/122;H04N19/186;H04N19/44;H04N19/60;H04N19/176 主分类号 G06K9/36
代理机构 Wenderoth, Lind & Ponack, L.L.P. 代理人 Wenderoth, Lind & Ponack, L.L.P.
主权项 1. An image decoding method of decoding an image that includes coding units, each of the coding units including a plurality of transform units, the image decoding method comprising: obtaining sets including luminance data and chrominance data from a coded bitstream on a per coding unit basis, the sets corresponding to the plurality of transform units, respectively, and the sets being obtained from the coded bitstream in order; decoding, for each of the plurality of transform units, luminance data and chrominance data included in the corresponding set; and performing, for each of the plurality of transform units, transformation on the decoded luminance data and the decoded chrominance data, wherein the coding units include a coding unit having a first transform unit and a second transform unit, wherein, in the obtaining, the set of the transformed luminance data and the transformed chrominance data corresponding to the first transform unit is obtained from the bitstream before the set of the transformed luminance data and the transformed chrominance data corresponding to the second transform unit is obtained from the bitstream, wherein the coding units are included in a slice of the image, wherein each of the coding units corresponds to a square region included in the slice, and wherein the luminance data and the chrominance data are grouped in the coded bitstream on a per predetermined block basis, the predetermined block is greater than or equal to a predetermined size, wherein when a transform unit in the predetermined block is smaller than the predetermined size (i) chrominance data of transform units in the predetermined block are combined into chrominance data for the transform units and (ii) the combined chrominance data is transformed, and wherein when the transform unit in the predetermined block is greater than or equal to the predetermined size, the chrominance data is transformed without combining.
地址 New York NY US