Method and apparatus for scaling an image in segments

1. A method for scaling an image in segments, the method comprising:

identifying scene features in a sequence of video frames;

determining a distribution of scene features in the sequence of video frames;

splitting a first video frame in the sequence of frames into a plurality of blocks;

assigning importance grade coefficients to the plurality of blocks in accordance with the distribution of scene features;

obtaining scaling coefficients for the plurality of blocks in accordance with the importance grade coefficients, wherein at least some of the blocks have different scale coefficients; and

scaling each block in the plurality of blocks in accordance with the scaling coefficients.

2. The method according to claim 1 , wherein scaling each block in the plurality of blocks in accordance with the scaling coefficients comprises:

applying a fourth-order 32-phase interpolation algorithm in a horizontal direction, and a second-order 32-phase interpolation algorithm in a vertical direction in each of the plurality of blocks according to the scaling coefficients;

using an interpolation filtering unit for the horizontal direction and the vertical direction; and

reusing a same interpolator on a time-sharing basis in the horizontal direction and the vertical direction to scale the image.

3. An apparatus for scaling an image in segments, the apparatus comprising:

a scene identifying unit, configured to identify a distribution of scene features in a sequence of video frames; and

a segmental scaling unit, configured to: split a first video frame in the sequence of frames into a plurality of blocks; assign importance grade coefficients to the plurality of blocks in accordance with the distribution of scene features; obtain scaling coefficients for the plurality of blocks in accordance with the importance grade coefficients; and scale each block in the plurality of blocks in accordance with the scaling coefficients.

4. The apparatus according to claim 3 , wherein the segmental scaling unit is configured to scale each block in the plurality of blocks by applying a fourth-order 32-phase interpolation algorithm in a horizontal direction and a second-order 32-phase interpolation algorithm in a vertical direction in each of the blocks according to the scaling coefficients; use an interpolation filtering unit uniformly; and reuse a same interpolator on a time-sharing basis in the horizontal direction and the vertical direction to scale the image.

5. The apparatus according to claim 3 , wherein the apparatus is implemented in hardware that is instructed by a program.

6. The apparatus according to claim 3 , wherein identifying the distribution of scene features in the video frame comprises using feature matching to identify the scene features in the sequence of video frames.

7. The apparatus according to claim 6 , wherein using feature matching to identify the scene features includes performing face detection on frames in the sequence of frames.

8. The apparatus according to claim 7 , wherein performing face detection includes utilizing a Haar detection technique.

9. The apparatus according to claim 3 , wherein assigning importance grade coefficients comprises:

assigning higher importance grade coefficients to blocks depicting portions of a human face than to blocks not depicting portions of a human face.

10. The apparatus according to claim 9 , wherein at least some blocks in the plurality of blocks have different importance grade coefficients.

11. The apparatus according to claim 10 , wherein blocks having different importance grade coefficients have different scaling coefficients.

12. The apparatus according to claim 3 , wherein the apparatus is implemented in hardware.

13. The method according to claim 1 , wherein identifying the distribution of scene features in the sequence of video frames comprises using feature matching to identify scene features in the sequence of video frames.

14. The method according to claim 1 , wherein using feature matching to identify the scene features includes performing face detection on frames in the sequence of frames.

15. The method according to claim 1 , wherein performing face detection includes utilizing a Haar detection technique.

16. The method according to claim 1 , wherein assigning importance grade coefficients comprises:

assigning higher importance grade coefficients to blocks depicting portions of a human face than to blocks not depicting portions of a human face.

17. The method according to claim 1 , wherein at least some of the blocks have different importance grade coefficients.

18. The method according to claim 17 , wherein blocks having different importance grade coefficients have different scaling coefficients.

19. The method according to claim 1 , wherein the method is implemented by a program instructing hardware.

20. An apparatus comprising:

a processor; and

a computer readable storage medium storing programming for execution by the processor, the programming including instructions to:

identify scene features in a sequence of video frames;

determine a distribution of scene features in the sequence of video frames;

split a first video frame in the sequence of frames into a plurality of blocks;

assign importance grade coefficients to the plurality of blocks in accordance with the distribution of scene features;

obtain scaling coefficients for the plurality of blocks in accordance with the importance grade coefficients; and

scale each block in the plurality of blocks in accordance with the scaling coefficients.

21. The apparatus according to claim 20 , wherein at least some of the blocks have different importance grade coefficients.

22. The apparatus according to claim 21 , wherein blocks having different importance grade coefficients have different scaling coefficients.