# US 2012/0250768 A1 · Video decoding switchable between two modes

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

- Number: US 2012/0250768 A1
- Document ID: US20120250768A1
- Status: Published application
- Kind: A1
- Application number: 13/429,089
- Filed: 2012-03-23
- Published: 2012-10-04
- Earliest claimed: 2011-04-04
- Assignees: NXP B.V.
- Inventors: Manivel Sethu; Francois Martin; Pradeep Muruganandam; Sudeendra Maddur Gundurao
- Web page: https://patentgrep.com/patent/US20120250768A1

## CPC

- H04N 19/122: Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
- H04N 19/132: Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
- H04N 19/154: Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion (use of rate-distortion criteria H04N19/147)
- H04N 19/164: Feedback from the receiver or from the transmission channel
- H04N 19/17: the unit being an image region, e.g. an object
- H04N 19/172: the region being a picture, frame or field
- H04N 19/428: {Recompression, e.g. by spatial or temporal decimation}
- H04N 19/43: Hardware specially adapted for motion estimation or compensation
- H04N 19/44: Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
- H04N 19/547: Motion estimation performed in a transform domain
- H04N 19/61: in combination with predictive coding
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

Video decoder apparatus and method, for decoding a motion-compensated transform-coded video stream. The apparatus has: a first mode in which it is operable to decode the video stream at a first resolution, and wherein a motion-compensation step of the decoding is performed in the image-domain. It also has a second mode in which it is operable to decode the video stream at a second resolution, and wherein at least one motion-compensation step is performed in the transform-domain. The decoder is adapted to switch between the two modes while decoding the video stream.

## Claims

### Claim 1

Video decoder apparatus for decoding a motion-compensated transform-coded video stream, the apparatus comprising:
a first mode in which it is operable to decode the video stream at a first resolution, and wherein a motion-compensation step of the decoding is performed in the image-domain; and
a second mode in which it is operable to decode the video stream at a second resolution, and wherein at least one motion-compensation step is performed in the transform-domain,
the decoder being adapted to switch between the two modes while decoding the video stream.

### Claim 2 (depends on 1)

The apparatus of claim 1, wherein the decoder is adapted to, upon switching from the first mode to the second mode, transform a reference frame for motion-compensation from the image-domain into the transform-domain.

### Claim 3 (depends on 1)

The apparatus of claim 1, wherein the decoder is operable, in the second mode, to perform a motion-compensation step for a B-frame in the image domain, and wherein
when the next frame to be decoded after switching to the second mode is a B-frame, the decoder is further adapted to:
transform a reference frame from the image-domain to the transform-domain at the first resolution; and
inverse transform the reference frame from the transform-domain back to the image-domain at the second resolution.

### Claim 4 (depends on 2)

The apparatus of claim 2, wherein the second resolution is lower than the first resolution and wherein transforming the reference frame from the image-domain to the transform-domain comprises calculating only a subset of the transform coefficients for the reference frame.

### Claim 5 (depends on 4)

The apparatus of claim 4, wherein the transform is a block-transform and the subset chosen is at least one of:
a contiguous sub-block of the block of coefficients in the transform-domain;
a contiguous subset of the coefficients including a coefficient at one corner of the block and being symmetrical about the main diagonal of the block; and
a contiguous subset of the coefficients, chosen according to a diagonal zig-zag pattern starting from one corner of the block.

### Claim 6 (depends on 1)

The apparatus of claim 1, wherein the decoder is adapted to, upon switching from the second mode to the first mode, inverse transform a reference frame for motion-compensation from the transform-domain into the image-domain.

### Claim 7 (depends on 6)

The apparatus of claim 6, wherein the decoder is adapted to inverse transform the reference frame into the image domain at the first resolution.

### Claim 8 (depends on 6)

The apparatus of claim 6, wherein the decoder is adapted to:
inverse transform the reference frame into the image domain at a third resolution different from the first resolution; followed by
resizing the frame to the first resolution.

### Claim 9 (depends on 8)

The apparatus of claim 8, wherein:
the third resolution is greater than the second resolution; and
inverse transforming the reference frame from the transform-domain to the image-domain comprises calculating pixel-values at the third resolution using only the subset of the transform-coefficients which is available at the second resolution.

### Claim 10

A method of decoding a video stream that has been encoded using motion-compensated transform-coding, the method comprising:
in a first mode, decoding the video stream at a first resolution, wherein a motion-compensation step of the decoding is performed in the image-domain; and
in a second mode, decoding the video stream at a second resolution, wherein at least one motion-compensation step of the decoding is performed in the transform-domain,
the method further comprising switching from one mode to the other while decoding the video stream.

### Claim 11 (depends on 10)

The method of claim 10, further comprising, upon switching from the first mode to the second mode, transforming a reference frame for motion-compensation from the image-domain into the transform-domain.

### Claim 12 (depends on 10)

The method of claim 10, wherein, in the second mode, the method comprises performing a motion-compensation step for a B-frame in the image domain, and wherein
when the next frame to be decoded after switching to the second mode is a B-frame, the method comprises:
transforming a reference frame from the image-domain to the transform-domain at the first resolution; and
inverse transforming the reference frame from the transform-domain back to the image-domain at the second resolution.

### Claim 13 (depends on 10)

The method of claim 10, further comprising, upon switching from the second mode to the first mode, inverse transforming a reference frame for motion-compensation from the transform-domain into the image-domain.

### Claim 14 (depends on 10)

A computer program comprising computer program code means adapted to perform all the steps of any of claim 10 when said program is run on a computer.

### Claim 15 (depends on 14)

A computer program as claimed in claim 14 embodied on a non-transitory computer readable medium.
