Trick play using temporal layering

1. A method for encoding video content into a temporal layered coding structure comprising:

defining a plurality of sub-sequence layers in a layer hierarchy;

selectively assigning each of a plurality of images within a sub-sequence of a group of images to a respective one of said sub-sequence layers such that a grouping of images contained in any sequential combination of sub-sequence layers that includes a hierarchically lowest of said sub-sequence layers results in a group of images that temporally are substantially evenly spaced;

wherein intra images (I-images) are exclusively assigned to said hierarchically lowest sub-sequence layer, predictive images (P-images) are assigned to a second of said sub-sequence layers, said second sub-sequence layer being one level higher than said lowest sub-sequence layer in said layer hierarchy and a selected first set of B R -images are assigned to a third of said sub-sequence layers, said third sub-sequence layer being one level higher than said second sub-sequence layer in said layer hierarchy, said B R -images being predicted exclusively from images selected from the group consisting of I-images and P-images.

2. The method of claim 1 , further comprising:

predicting each non-intra image exclusively from at least one of said plurality of images which are assigned to a sub-sequence layer that is equal or lower in said layer hierarchy than the sub-sequence layer containing said each non-intra image.

3. The method according to claim 1 , wherein said I-images temporally are substantially evenly spaced.

4. The method of claim 1 , wherein said I-images and said P-images form a group of images that temporally are substantially evenly spaced.

5. The method of claim 1 , wherein said I-images, said P-images and said BR-images form a group of images that temporally are substantially evenly spaced.

6. The method of claim 1 , further comprising:

selecting a second set of bidirectional images (B U -images), said B U -images being predicted exclusively from images selected from the group consisting of I-images, P-images and B R -images;

assigning said B U -images to a fourth of said sub-sequence layers, said fourth sub-sequence layer being one level higher than said third sub-sequence layer in said layer hierarchy.

7. The method of claim 1 , further comprising providing sub-sequence layer hierarchy structural information in at least one message selected from the group consisting of sub-sequence information (SEI) messages, sub-sequence layer characteristic SEI massages and video usability information (VUI) messages, said layer hierarchy structural information indicating the temporal layered coding structure.

8. The method of claim 6 , wherein said B U -images temporally are substantially evenly spaced in said fourth sub-sequence layer.

9. A video encoder comprising a processor and a memory, for encoding video content into a temporal layered coding structure, said video encoder adapted to define a plurality of sub-sequence layers in a layer hierarchy and selectively assign each of a plurality of images within a sub-sequence of a group of images to a respective one of said sub-sequence layers such that a grouping of images contained in any sequential combination of sub-sequence layers that includes a hierarchically lowest of said sub-sequence layers results in a group of images that temporally are substantially evenly spaced;

wherein said video encoder assigns intra images (I-images) exclusively to said hierarchically lowest sub-sequence layer and predictive images (P-images) to a second of said sub-sequence layers, said second sub-sequence layer being one level higher than said lowest sub-sequence layer in said layer hierarchy and a selected first set of B R -images are assigned to a third of said sub-sequence layers, said third sub-sequence layer being one level higher than said second sub-sequence layer in said layer hierarchy, said B R -images being predicted exclusively from images selected from the group consisting of I-images and P-images.

10. The video encoder of claim 9 wherein said video encoder predicts each non-intra image exclusively from at least one of said plurality of images which are assigned to a sub-sequence layer that is equal or lower in said layer hierarchy than the sub-sequence layer containing said each non-intra image.

11. The video encoder of claim 9 , wherein said I-images temporally are substantially evenly spaced.

12. The video encoder of claim 9 , wherein said I-images and said P-images form a group of images that temporally are substantially evenly spaced.

13. The video encoder of claim 9 , wherein said I-images, said P-images and said B R -images form a group of images that temporally are substantially evenly spaced.

14. The video encoder of claim 9 wherein said video encoder selects a second set of bidirectional images (B U -images), said B U -images being predicted exclusively from images selected from the group consisting of I-images, P-images and B R -images, and said video encoder assigns said B U -images to a fourth of said sub-sequence layers, said fourth sub-sequence layer being one level higher than said third sub-sequence layer in said layer hierarchy.

15. The video encoder of claim 14 , wherein said B U -images temporally are substantially evenly spaced in said fourth sub-sequence layer.

16. The video encoder of claim 9 , comprising a message generator that provides sub-sequence layer hierarchy structural information in at least one message selected from the group consisting of sub-sequence information (SEI) messages, sub-sequence layer characteristic SEI massages and video usability information (VUI) messages, said layer hierarchy structural information indicating the temporal layered coding structure.