Granted Patent B2

US 12,309,370 · App. 18/390,668 · Granted May 20, 2025

Inventors: Masato Ohkawa (Toyama, JP); Hideo Saitou (Ishikawa, JP); Tadamasa Toma (Osaka, JP); Takahiro Nishi (Nara, JP); Kiyofumi Abe (Osaka, JP); Ryuichi Kanoh (Osaka, JP)

Assignee: PANASONIC INTELLECTUAL PROPERTY CORPORATION OF AMERICA

H04N19/12H04N19/124H04N19/18H04N19/625H04N19/619

Abstract

Claims (19)

1. An encoder that encodes a first current block and a second current block in a picture, the encoder comprising:

circuitry; and

memory, wherein

using the memory, the circuitry:

when an intra prediction is used for the first current block and the second current block,

performs a first transform on (i) a first residual signal of the first current block using a first transform basis and (ii) a second residual signal of the second current block using a second transform basis, to generate first transform coefficients;

performs a second transform on the first transform coefficients of the first current block using a third transform basis to generate second transform coefficients and quantizes the second transform coefficients, when the first transform basis used for the first transform on the first residual signal is same as a predetermined transform basis; and

quantizes the first transform coefficients of the second current block without performing the second transform, when the second transform basis used for the first transform on the second residual signal is different from the predetermined transform basis, and

when an inter prediction is used for the first current block and the second current block,

performs the first transform on the first residual signal and the second residual signal using a fourth transform basis to generate third transform coefficients and quantizes the third transform coefficients without performing the second transform regardless of whether or not the fourth transform basis is same as the predetermined transform basis.

2. A decoder that decodes a first current block and a second current block in a picture, the decoder comprising:

circuitry; and

memory, wherein

using the memory, the circuitry:

when an intra prediction is used for the first current block and the second current block,

(i) inverse quantizes quantized coefficients of the first current block to generate second transform coefficients, performs a second inverse transform on the second transform coefficients using a second inverse transform basis to generate first transform coefficients, and performs a first inverse transform on the first transform coefficients using a first inverse transform basis to generate a first residual signal of the first current block, when the first inverse transform basis is same as a predetermined inverse transform basis; and

(ii) inverse quantizes quantized coefficients of the second current block to generate third transform coefficients and performs the first inverse transform on the third transform coefficients using a third inverse transform basis to generate a second residual signal of the second current block without performing the second inverse transform, when the third inverse transform basis is different from the predetermined inverse transform basis, and

when an inter prediction is used for the first current block and the second current block,

inverse quantizes quantized coefficients of the first current block and the second current block to generate fourth transform coefficients and performs the first inverse transform on the fourth transform coefficients using a fourth inverse transform basis to generate the first residual signal and the second residual signal without performing the second inverse transform regardless of whether or not the fourth transform basis is same as the predetermined transform basis.

Continuity (5)

References Cited (45)