# US 2008/0186404 A1 · Video Error Concealment

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

- Number: US 2008/0186404 A1
- Document ID: US20080186404A1
- Status: Published application
- Kind: A1
- Application number: 11/794,922
- Filed: 2006-01-06
- Published: 2008-08-07
- Earliest claimed: 2005-01-08
- Assignees: Provision Communication Technologies Limited
- Inventors: David Roger Bull; Dimitris Agrafiotis
- Web page: https://patentgrep.com/patent/US20080186404A1

## CPC

- H04N 19/583: Motion compensation with overlapping blocks
- H04N 19/139: Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
- H04N 19/176: the region being a block, e.g. a macroblock
- H04N 19/593: involving spatial prediction techniques
- H04N 19/895: in combination with error concealment
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A temporal error concealment is described. A video stream comprises a current frame comprising a lost image region, and a reference frame. The method comprises forming a list of motion vector candidates, the list of motion vector candidates including a set of one or more motion vectors from the current frame and a set of one or more motion vectors from the reference frame; testing each of the motion vector candidates using a matching error measure; selecting a replacement motion vector from the list based on the matching error measures; and using the selected replacement motion vector to conceal the lost image region. A mode selection method is also described. A motion vector is used to identify a motion compensated reference image region in the reference frame. A motion compensated temporal activity of an image region in the current frame is calculated based on a comparison between the image region in the current frame and the motion compensated reference image region in the reference frame. A spatial or temporal error concealment mode is then selected based on the calculated temporal activity.

## Claims

### Claim at position 1

-14. (canceled)

### Claim 15

A method of concealing an error in a video stream, the video stream comprising a current frame comprising a lost image region, and a reference frame, the method comprising using a motion vector to identify a reference image region in the reference frame;
calculating a motion compensated temporal activity of an image region in the current frame based on a comparison between the image region in the current frame and the reference image region in the reference frame; selecting a spatial or temporal error concealment mode based on the calculated temporal activity; and concealing the lost image region using the selected error concealment mode.

### Claim 16 (depends on 15)

The method of claim 15 further comprising calculating a spatial activity of the image region in the current frame, and selecting a spatial or temporal error concealment mode based on the calculated temporal activity and the calculated spatial activity.

### Claim 17 (depends on 15)

The method of claim 15 further comprising calculating the motion vector by forming a list of motion vector candidates; testing each of the motion vector candidates using a matching error measure; and selecting a replacement motion vector from the list based on the matching error measures.

### Claim 18 (depends on 17)

The method of claim 17 wherein the list of motion vector candidates includes a set of one or more motion vectors from the current frame and a set of one or more motion vectors from the reference frame.

### Claim 19

A method of concealing an error in a video stream, the video stream comprising a current frame comprising a lost image region, and a reference frame, the method comprising forming a list of motion vector candidates, the list of motion vector candidates including a set of one or more motion vectors from the current frame and a set of one or more motion vectors from the reference frame; testing each of the motion vector candidates using a matching error measure; selecting a replacement motion vector from the list based on the matching error measures; and using the selected replacement motion vector to conceal the lost image region.

### Claim 20 (depends on 19)

The method of claim 19 further comprising performing motion refinement by forming a list of motion vector candidates using the selected replacement motion vector as a starting point; testing each of the motion vector candidates in the list using a matching error measure; selecting a motion-refined replacement motion vector from the list based on the matching error measures; and using the selected motion-refined replacement motion vector to conceal the lost image region.

### Claim 21 (depends on 19)

The method of claim 19 wherein the list of motion vector candidates comprises a set of motion vectors each associated with a neighbouring image region in the current frame which neighbours the lost image region.

### Claim 22 (depends on 19)

The method of claim 19 wherein the list of motion vector candidates comprises a motion vectors associated with a collocated image region in the reference frame, the collocated image region being collocated with the lost image region.

### Claim 23 (depends on 19)

The method of claim 19 wherein the list of motion vector candidates comprises a set of motion vectors each associated with a neighbouring image region in the reference frame which neighbours a collocated image region in the reference frame, the collocated image region being collocated with the lost image region.

### Claim 24 (depends on 19)

The method of claim 19 further comprising dividing the lost image region into multiple subregions; forming multiple replacement signals for each of the subregions using the selected replacement motion vector and the motion vector of one or more neighbouring subregions; blending the multiple replacement signals to generate a blended replacement signal; and using the blended replacement signal to conceal the lost image region.

### Claim 25 (depends on 19)

The method of claim 19 wherein the matching error measure for each motion vector candidate is calculated by calculating a sum of absolute differences between a pixel boundary adjacent to the lost image region and a pixel boundary adjacent to a replacement image region in a reference frame identified by the motion vector candidate.

### Claim 26 (depends on 15)

The method of claim 15 wherein the current frame is an intra-coded frame.

### Claim 27 (depends on 15)

A video decoder configured to decode a video stream and conceal an error in the video stream by the method of claim 15.

### Claim 28 (depends on 15)

Computer software configured to conceal an error in a video stream by the method of claim 15.

### Claim 29 (depends on 17)

The method of claim 17 further comprising performing motion refinement by forming a list of motion vector candidates using the selected replacement motion vector as a starting point; testing each of the motion vector candidates in the list using a matching error measure; selecting a motion-refined replacement motion vector from the list based on the matching error measures; and using the selected motion-refined replacement motion vector to conceal the lost image region.

### Claim 30 (depends on 17)

The method of claim 17 wherein the list of motion vector candidates comprises a set of motion vectors each associated with a neighbouring image region in the current frame which neighbours the lost image region.

### Claim 31 (depends on 17)

The method of claim 17 wherein the list of motion vector candidates comprises a motion vectors associated with a collocated image region in the reference frame, the collocated image region being collocated with the lost image region.

### Claim 32 (depends on 17)

The method of claim 17 wherein the list of motion vector candidates comprises a set of motion vectors each associated with a neighbouring image region in the reference frame which neighbours a collocated image region in the reference frame, the collocated image region being collocated with the lost image region.

### Claim 33 (depends on 17)

The method of claim 17 further comprising dividing the lost image region into multiple subregions; forming multiple replacement signals for each of the subregions using the selected replacement motion vector and the motion vector of one or more neighbouring subregions; blending the multiple replacement signals to generate a blended replacement signal; and using the blended replacement signal to conceal the lost image region.

### Claim 34 (depends on 17)

The method of claim 17 wherein the matching error measure for each motion vector candidate is calculated by calculating a sum of absolute differences between a pixel boundary adjacent to the lost image region and a pixel boundary adjacent to a replacement image region in a reference frame identified by the motion vector candidate.
