发明名称 |
DATA CONVERTING APPARATUS, DATA CONVERTING METHOD, LEARNING APPARATUS, LEANING METHOD, PROGRAM, AND RECORDING MEDIUM |
摘要 |
<p>The present invention relates to a data conversion apparatus and a learning device in which an image can be converted into a higher-quality image. A class-tap generating circuit (2) and a predictive-tap generating circuit (3) generate, from an SD image, a predictive tap used for determining the HD pixel of a specified block of an HD image. A classification circuit (4) classifies the HD pixel of the specified block based on the class tap. A coefficient RAM (7) obtains a tap coefficient for the class of the HD pixel of the specified block from tap coefficients obtained by learning the relationship between supervisor data and learner data for each class of at least one class by giving a predetermined constraint condition to the supervisor data. A predictive-computation circuit (8) and a decoding circuit (9) determine the HD pixel of the specified block by using the tap coefficient and the predictive tap. <IMAGE></p> |
申请公布号 |
EP1501296(A1) |
申请公布日期 |
2005.01.26 |
申请号 |
EP20030717683 |
申请日期 |
2003.04.22 |
申请人 |
SONY CORPORATION |
发明人 |
KONDO, TETSUJIRO |
分类号 |
G06F17/10;G06N3/00;G06T3/40;H04N7/01;(IPC1-7):H04N7/01;H04N1/387 |
主分类号 |
G06F17/10 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|