Field Prediction

Prediction for P pictures is made from the two most recently decoded reference fields. The simplest case is shown in Figure 13.15, used when predicting the first picture of a frame or when using field prediction within a frame.

Figure 13.15: P Picture Prediction of First Field or Field Prediction in a Frame Picture.

Predicting the second field of a frame also requires the two most recently decoded reference fields. This is shown in Figure 13.16 where the second picture is the bottom field and in Figure 13.17 where the second picture is the top field.

Figure 13.16: P Picture Prediction of Second Field Picture (Bottom Field).

Figure 13.17: P Picture Prediction of Second Field Picture (Top Field).

Field prediction for B pictures is made from the two fields of the two most recent reference frames, as shown in Figure 13.18.

Figure 13.18: Field Prediction of B Field or Frame Pictures.

Frame Prediction

Prediction for P pictures is made from the most recently decoded picture, as shown in Figure 13.19. The reference picture may have been coded as either two fields or a single frame.

Figure 13.19: Frame Prediction For P Pictures.

Frame prediction for B pictures is made from the two most recent reference frames, as shown in Figure...