Adaptive motion search range

• US 20070064803A1
• Filed: 09/16/2005
• Published: 03/22/2007
• Est. Priority Date: 09/16/2005
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Original Assignee: SONY CORPORATION AND SONY ELECTRONICS INC.
• Inventors: Paniconi, Marco, Miao, Zhourong, Carrig, James
• Granted Patent: US 8,107,748 B2
• Time in Patent Office: Days
• US Class Current: 375/240.160
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/56 Motion estimation with initialisation of the vector search, e.g. estimating a good candidate to initiate a search
• CPC Class Codes: H04N 19/57 Motion estimation characterised by a search window with variable size or shape

Abstract

A technique for performing motion estimation using an adaptive motion search range includes calculating motion vectors for blocks of pixels in a target image, using the calculated motion vectors to generate a search range associated with a target block of pixels in the target image, and using the generated search range to estimate motion of the target block of pixels.

Claims

1. A method carried out by an electronic data processor, comprising:

calculating motion vectors for blocks of pixels in a target image;

using the calculated motion vectors to generate a search range associated with a target block of pixels in the target image; and

using the generated search range to estimate motion of the target block of pixels.

2. The method of claim 1, wherein generating the search range comprises:

using the calculated motion vectors to determine an estimated motion range.

3. The method of claim 2, wherein determining the estimated motion range comprises:

applying a low pass filter to the calculated motion vectors.

4. The method of claim 3, wherein applying the filter to the calculated motion vectors comprises:

{circumflex over (mx)}_i,j=L×

(M_x)^T, {circumflex over (my)}_i,j=L×

(M_y)^T,

where L is a filter having n taps, such that L=[l₁, l₂, . . . , l_n], and l_i is the filter coefficient of tap i;

{circumflex over (mx)}_i,j is the x-axis value of the estimated motion range of the target block;

M_x is a vector containing x-axis values of the calculated motion vectors;

(M_x)^T is a transpose of vector M_x;

{circumflex over (my)}_i,j is the y-axis value of the estimated motion range of the target block;

M_y is a vector containing y-axis values of the calculated motion vectors; and

(M_y)^T is a transpose of vector M_y.

5. The method of claim 1, wherein generating the motion search range comprises:

determining a variance of the calculated motion vectors.

6. The method of claim 5, wherein determining the variance comprises:

applying a high pass filter to the calculated motion vectors.

7. The method of claim 1, further comprising:

determining a block scanning order for the blocks in the target image selected from the group consisting of;

a raster scanning order and a sub-sampling order.

8. The method of claim 1, wherein calculating motion vectors for blocks of pixels in the target image comprises:

calculating the motion vectors using multiple reference images.

9. An apparatus comprising:

a motion vector calculator that calculates motion vectors for blocks of pixels in a target image;

a search range calculator that uses the calculated motion vectors to generate a search range associated with a target block of pixels in the target image; and

a motion estimator that uses the generated search range to estimate motion of the target block of pixels.

10. The apparatus of claim 9, wherein the search range calculator uses the calculated motion vectors to determine an estimated motion range.

11. The apparatus of claim 10, wherein the search range calculator applies a low pass filter to the calculated motion vectors.

12. The apparatus of claim 11, wherein the search range calculator applies

{circumflex over (mx)}_i,j=L×

(M_x)^T, {circumflex over (my)}_i,j=L×

(M_y)^T,

where L is a filter having n taps, such that L=[l₁, l₂, . . . , l_n], and l_i is the filter coefficient of tap i;

{circumflex over (mx)}_i,j is the x-axis value of the estimated motion range of the target block;

M_x is a vector containing x-axis values of the calculated motion vectors;

(M_x)^T is a transpose of vector M_x;

{circumflex over (my)}_i,j is the y-axis value of the estimated motion range of the target block;

M_y is a vector containing y-axis values of the calculated motion vectors; and

(M_y)^T is a transpose of vector M_y.

13. The apparatus of claim 9, wherein the search range calculator determines a variance of the calculated motion vectors.

14. The apparatus of claim 13, wherein the search range calculator determines the variance by applying a high pass filter to the calculated motion vectors.

15. The apparatus of claim 9, wherein the blocks in the target image are scanned in a raster scanning order or a sub-sampling order.

16. The apparatus of claim 9, wherein the motion vector calculator calculates the motion vectors using multiple reference images.

17. A computer readable medium storing a program of instructions which, when executed by a processing system, cause the system to perform a method comprising:

calculating motion vectors for blocks of pixels in a target image;

using the calculated motion vectors to generate a search range associated with a target block of pixels in the target image; and

using the generated search range to estimate motion of the target block of pixels.

18. The computer readable medium of claim 17, wherein generating the search range comprises:

using the calculated motion vectors to determine an estimated motion range.

19. The computer readable medium of claim 18, wherein determining the estimated motion range comprises:

applying a low pass filter to the calculated motion vectors.

20. The computer readable medium of claim 19, wherein applying the filter to the calculated motion vectors comprises:

{circumflex over (mx)}_i,j=L×

(M_x)^T, {circumflex over (my)}_i,j=L×

(M_y)^T,

where L is a filter having n taps, such that L=[l₁, l₂, . . . , l_n], and l_i is the filter coefficient of tap i;

{circumflex over (mx)}_i,j is the x-axis value of the estimated motion range of the target block;

M_x is a vector containing x-axis values of the calculated motion vectors;

(M_x)^T is a transpose of vector M_x;

{circumflex over (my)}_i,j is the y-axis value of the estimated motion range of the target block;

M_y is a vector containing y-axis values of the calculated motion vectors; and

(M_y)^T is a transpose of vector M_y.

21. The computer readable medium of claim 17, wherein generating the motion search range comprises:

determining a variance of the calculated motion vectors.

22. The computer readable medium of claim 21, wherein determining the variance comprises:

applying a high pass filter to the calculated motion vectors.

23. The computer readable medium of claim 17, further comprising:

determining a block scanning order for the blocks in the target image selected from the group consisting of;

a raster scanning order and a sub-sampling order.

24. The computer readable medium of claim 17, wherein calculating motion vectors for blocks of pixels in the target image comprises:

calculating the motion vectors using multiple reference images.