# US 2020/0404296 A1 · Method and a device for picture encoding and decoding

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2020/0404296 A1
- Document ID: US20200404296A1
- Status: Published application
- Kind: A1
- Application number: 16/490,589
- Filed: 2018-02-21
- Published: 2020-12-24
- Earliest claimed: 2017-03-02
- Assignees: InterDigital VC Holdings, Inc.
- Applicants: Interdigital Vc Holdings, Inc.
- Inventors: Saurabh PURI; Sebastien LASSERRE; Patrick LE CALLET
- Web page: https://patentgrep.com/patent/US20200404296A1

## CPC

- H04N 19/184: the unit being bits, e.g. of the compressed video stream
- H04N 19/70: characterised by syntax aspects related to video coding, e.g. related to compression standards
- H04N 19/12: Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
- H04N 19/147: according to rate distortion criteria (rate-distortion as a criterion for motion estimation H04N19/567)
- H04N 19/176: the region being a block, e.g. a macroblock
- H04N 19/18: the unit being a set of transform coefficients
- H04N 19/44: Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
- H04N 19/60: using transform coding
- H04N 19/91: Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A decoding method is disclosed. The decoding method comprises: determining an index from a bitstream, the index identifying a coding element in a set of a plurality of coding elements; decoding a picture block from the bitstream using the coding element of the set identified by the index. In a specific embodiment, determining the index comprises: determining, for each coding element in the set, a probability that the coding element was used for encoding the picture block; and determining the index responsive to the determined probabilities.

## Claims

### Claim 1

A decoding method comprising:
determining an index from a bitstream, said index identifying a coding element in a set of a plurality of coding elements;
decoding a picture block from the bitstream using the coding element of the set identified by the index;
wherein determining said index comprises:
determining, for each coding element in said set, a probability that said coding element was used for encoding said picture block; and
determining said index responsive to said determined probabilities.

### Claim 2 (depends on 1)

The decoding method according to claim 1, wherein determining, for each coding element in said set, a probability comprises applying a convolutional neural network model on data decoded from the bitstream prior to determination of said index.

### Claim 3 (depends on 1)

The decoding method according to claim 1, wherein determining said index responsive to said determined probabilities comprises sorting the indices of the plurality of coding elements in a decreasing order of their probability to obtain a sorted list of indices, obtaining a truncated unary code from said bitstream and determining said index as the index of said sorted list of indices identified by said truncated unary code.

### Claim 4 (depends on 3)

The decoding method according to claim 3, wherein obtaining the truncated unary code comprises entropy decoding a part of said bitstream into said truncated unary code.

### Claim 5 (depends on 1)

The decoding method according to claim 1, wherein said coding element is a transform.

### Claim 6

An encoding method comprising:
encoding a picture block in a bitstream using a coding element of a set of a plurality of coding elements;
encoding an index identifying the coding element in the set of a plurality of coding elements in the bitstream;
wherein encoding said index comprises:
determining, for each coding element in said set, a probability that said coding element was used for encoding said picture block; and
encoding said index responsive to said determined probabilities.

### Claim 7 (depends on 6)

The encoding method according to claim 6, wherein determining, for each coding element in said set, a probability comprises applying a convolutional neural network model on data encoded in the bitstream prior to said index.

### Claim 8 (depends on 6)

The encoding method according to claim 6, wherein encoding said index responsive to said determined probabilities comprises sorting the indices of the plurality of coding elements in a decreasing order of their probability to obtain a sorted list of indices, determining a truncated unary code identifying said index in said sorted list of indices and encoding said truncated unary code in said bitstream.

### Claim 9 (depends on 6)

The encoding method according to claim 6, wherein said coding element is a transform.

### Claim 10

A decoding device comprising:
means for determining an index from a bitstream, said index identifying a coding element in a set of a plurality of coding elements;
means for decoding a picture block from the bitstream using the coding element of the set identified by the index;
wherein said means for determining said index comprises:
means for determining, for each coding element in said set, a probability that said coding element was used for encoding said picture block; and
means for determining said index responsive to said determined probabilities.

### Claim 11 (depends on 10)

The decoding device according to claim 10, wherein said means for determining, for each coding element in said set, a probability comprises means for applying a convolutional neural network model on data decoded from the bitstream prior to determination of said index.

### Claim 12 (depends on 10)

The decoding device according to claim 10, wherein said means for determining said index responsive to said determined probabilities comprises means for sorting the indices of the plurality of coding elements in a decreasing order of their probability to obtain a sorted list of indices, means for obtaining a truncated unary code from said bitstream and means for determining said index as the index of said sorted list of indices identified by said truncated unary code.

### Claim 13 (depends on 12)

The decoding device according to claim 12, wherein said means for obtaining the truncated unary code comprises means for entropy decoding a part of said bitstream into said truncated unary code.

### Claim 14 (depends on 10)

The decoding device according to claim 10, wherein said coding element is a transform.

### Claim 15

An encoding device comprising:
means for encoding a picture block in a bitstream using a coding element of a set of a plurality of coding elements;
means for encoding an index identifying the coding element in the set of a plurality of coding elements in the bitstream;
wherein said means for encoding said index comprises:
means for determining, for each coding element in said set, a probability that said coding element was used for encoding said picture block; and
means for encoding said index responsive to said determined probabilities.

### Claim 16 (depends on 15)

The encoding device according to claim 15, wherein said means for determining, for each coding element in said set, a probability comprises means for applying a convolutional neural network model on data encoded in the bitstream prior to said index.

### Claim 17 (depends on 15)

The encoding device according to claim 15, wherein said means for encoding said index responsive to said determined probabilities comprises means for sorting the indices of the plurality of coding elements in a decreasing order of their probability to obtain a sorted list of indices, means for determining a truncated unary code identifying said index in said sorted list of indices and means for encoding said truncated unary code in said bitstream.

### Claim 18 (depends on 15)

The encoding device according to claim 15, wherein said coding element is a transform.
