# US 2010/0166072 A1 · Motion Estimation Techniques

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

- Number: US 2010/0166072 A1
- Document ID: US20100166072A1
- Status: Published application
- Kind: A1
- Application number: 12/346,979
- Filed: 2008-12-31
- Published: 2010-07-01
- Earliest claimed: 2008-12-31
- Assignees: Intel Corporation
- Inventors: Dov Bukin
- Web page: https://patentgrep.com/patent/US20100166072A1

## CPC

- H04N 19/567: Motion estimation based on rate distortion criteria
- H04N 19/533: Motion estimation using multistep search, e.g. two-dimensional [2D]-log search or one-at-a-time search [OTS]
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

Techniques for motion estimation are disclosed. For example, embodiments may determine error metric lower bound values for a block in a current image, where each of the error metric lower bound values corresponds to a search position in a reference image. Based on a separation threshold, a plurality of candidate positions may be identified from the search positions. A matching position is then identified from the candidate positions. Based on the matching position, a motion vector is determined.

## Claims

### Claim 1

A method, comprising:
determining a plurality of error metric lower bound values for a block in a current image, wherein each of the error metric lower bound values corresponds to a search position in a reference image;
selecting a plurality of candidate positions from the plurality of search positions based on a separation threshold;
identifying one of the candidate positions as a matching position for the block in the current image; and
determining a motion vector based on the block in the current image and the matching position.

### Claim 2 (depends on 1)

The method of claim 1, wherein each of the error metric lower bound values is a subsampled sum of absolute differences (SSAD) value.

### Claim 3 (depends on 1)

The method of claim 1, wherein said identifying comprises calculating an error metric for each of the candidate positions; and
selecting a minimum of the error metrics.

### Claim 4 (depends on 3)

The method of claim 3, wherein the plurality of candidate positions includes a zero displacement position.

### Claim 5 (depends on 3)

The method of claim 3, wherein each of the error metrics is a sum of absolute differences (SAD) value.

### Claim 6 (depends on 3)

The method of claim 3, further comprising determining the separation threshold.

### Claim 7 (depends on 1)

The method of claim 1, further comprising encoding the current image based at least on the motion vector.

### Claim 8

An apparatus, comprising:
a storage medium to store pixel data for a current image and a reference image;
a block selection module to select a block within the current image; and
a vector calculation module to:
determine a plurality of error metric lower bound values for the selected block, wherein each of the error metric lower bound values corresponds to a search position in the reference image;
select a plurality of candidate positions from the plurality of search positions based on a separation threshold;
identify one of the candidate positions as a matching position for the block in the current image; and
determine a motion vector based on the block in the current image and the matching position.

### Claim 9 (depends on 8)

The apparatus of claim 8, wherein each of the error metric lower bound values is a subsampled sum of absolute differences (SSAD) value.

### Claim 10 (depends on 8)

The apparatus of claim 8, wherein each of the error metrics is a sum of absolute differences (SAD) value.

### Claim 11 (depends on 8)

The apparatus claim 8, wherein the vector calculation module further determines the separation threshold.

### Claim 12

An apparatus, comprising:
A storage medium to store pixel data for a current image and a reference image;
an encoder to encode the current image, wherein the encoder includes a motion estimation module to generate a motion vector for a block within the image; and
wherein the motion estimation module includes a vector calculation module to:
determine a plurality of error metric lower bound values for the selected block, wherein each of the error metric lower bound values corresponds to a search position in the reference image;
select a plurality of candidate positions from the plurality of search positions based on a separation threshold;
identify one of the candidate positions as a matching position for the block in the current image; and
determine the motion vector based on the block in the current image and the matching position.

### Claim 13 (depends on 12)

The apparatus of claim 12, wherein each of the error metric lower bound values is a subsampled sum of absolute differences (SSAD) value.

### Claim 14 (depends on 12)

The apparatus of claim 12, wherein each of the error metrics is a sum of absolute differences (SAD) value.

### Claim 15 (depends on 12)

The apparatus claim 12, wherein the vector calculation module further determines the separation threshold.

### Claim 16

An article comprising a machine-accessible medium having stored thereon instructions that, when executed by a machine, cause the machine to:
determine a plurality of error metric lower bound values for a block in a current image, wherein each of the error metric lower bound values corresponds to a search position in a reference image;
select a plurality of candidate positions from the plurality of search positions based on a separation threshold;
identify one of the candidate positions as a matching position for the block in the current image; and
determine a motion vector based on the block in the current image and the matching position.

### Claim 17 (depends on 16)

The article of claim 16, wherein the instructions, when executed by a machine, further cause the machine to determine the separation threshold.

### Claim 18 (depends on 16)

The article of claim 16, wherein each of the error metric lower bound values is a subsampled sum of absolute differences (SSAD) value.

### Claim 19 (depends on 16)

The article of claim 16, wherein each of the error metrics is a sum of absolute differences (SAD) value.

### Claim 20 (depends on 16)

The article of claim 16, wherein the plurality of candidate positions includes a zero displacement position.
