Video encoder with embedded scene change and 3:2 pull-down detections

1. A method for encoding a video sequence comprising:

executing a first phase of motion estimation to determine a set of field motion vectors describing a relationship between fields of same polarity in two frames;

executing a scene change detection using the set of field motion vectors;

executing a 3:2 pulldown detection using the set of field motion vectors if no scene change is detected; and

executing a second phase of motion estimation to determine a set of motion vectors describing a relationship between fields of opposite polarity in the two frames and a relationship between the two frames.

2. The method of claim 1 , wherein

the set of field motion vectors comprises a first set of motion vectors between a first field of a first frame and a first field of a second frames, and a second set of motion vectors between a second field of the first frame and a second field of the second frame.

3. The method of claim 2 , wherein the set of motion vectors comprises:

a third set of motion vectors between the first field of the first frame and the second field of the second frame;

a fourth set of motion vectors between the second field of the first frame and the first field of the second frame; and

a fifth set of motion vectors between the first frame and the second frame.

4. The method of claim 1 , further comprising:

if the 3:2 pulldown detection detects a repeated field, removing the repeated field.

5. A video encoder comprising:

a motion detection component having a first phase and a second phase, the first phase to determine first and second set of motion vectors describing a relationship between fields of same polarity in two frames, and the second phase to determine third and fourth sets of motion vectors describing a relationship between fields of opposite polarity in the two frames, and to determine a fifth set of motion vectors describing a relationship between the two frames;

a scene change detection component to detect a scene change using the first and second set of motion vectors; and

a 3:2 pulldown detection component to detect a repeated field using the first and second set of motion vectors if no scene change is detected.

6. The video encoder of claim 5 , wherein the first set of motion vectors comprises motion vectors between a first field of a first frame and a first field of a second frame, and the second set of motion vectors comprises motion vectors between a second field of the first frame and a second field of the second frame.

7. The video encoder of claim 5 , wherein the scene change detection component detect a scene change by comparing a ratio of the first and second motion vectors to a threshold.

8. The video encoder of claim 5 , wherein the 3:2 pulldown detection component detect a repeated field by comparing a ratio of the first and second motion vectors to a threshold.

9. The video encoder of claim 8 further comprising a 3:2 pulldown undo component to compensate for finding a repeated field.

10. The video encoder of claim 9 , wherein the 3:2 pulldown undo component compensates for finding a repeated field by replacing the repeated field with a reference to a field from which the repeated field is repeated.

11. The video encoder of claim 9 , wherein the 3:2 pulldown undo component compensates for finding a repeated field by averaging the repeated field and a field from which the repeated field is repeated.

12. The video encoder of claim 5 , wherein the encoder is embodied in a processor.

13. A computer readable medium storing executable computer program instructions which, when executed by a processor, cause the processor to perform a method comprising:

executing a first phase of motion estimation, the first phase determining a set of field motion vectors describing a relationship between fields of same polarity in two frames;

executing a scene change detection using the set of field motion vectors;

executing 3:2 pulldown detection using the set of field motion vectors if no scene change is detected; and

executing second phase of motion estimation to determine a set of motion vectors describing a relationship between fields of opposite polarity in the two frames and a relationship between the two frames.

14. The computer medium of claim 13 , wherein

the set of field motion vectors comprises a first set of motion vectors between a first field of a first frame and a first field of a second frame and a second set of motion vectors between a second field of the first frame and a second field of the second frame.

15. The computer readable medium of claim 13 , wherein the set of motion vectors comprises:

a third set of motion vectors between the first field of the first frame and the second field of the second frame;

a fourth set of motion vectors between the second field of the first frame and the first field of the second frame; and

a fifth set of motion vectors between the first frame and the second frame.

16. The computer readable medium of claim 13 , further comprising:

if the 3:2 pull-down detection detects a repeated field, removing the repeated field.

17. An encoding apparatus comprising:

means for estimating motion comprising first and second phase means, the first phase means for determining a set of field motion vectors describing a relationship between fields of same polarity in two frames and the second phase means for determining a set of motion vectors describing a relationship between fields of opposite polarity in the two frames and a relationship between the two frames;

means to detect a scene change using the set of field motion vectors; and

means to detect a repeated field using the set of field motion vectors if no scene change is detected.