Multiple reference layer prediction signaling techniques
The disclosed subject matter, in one embodiment, provides techniques to signal inter-layer texture and motion prediction from different direct dependent reference layers. In certain exemplary arrangements, techniques are provided which include one or more syntax elements in a high level syntax structure, e.g., the slice segment header, indicating such different direct dependent reference layer(s).
1. A method for decoding in a decoding device a bitstream comprising at least three layers or views, each including at least one picture P0, P1 and P2, respectively, wherein, P2 is predicted at least in part from P1, and P1 is predicted at least in part from P0 without any data indicating explicit inter-layer prediction data in the bitstream, wherein P2 is predicted at least in part from P0 in the presence of explicit inter-layer prediction, the method comprising:
decoding with a decoder a slice segment header of the picture P2; and
reconstructing at least one sample of the picture P2 using information from the picture P0 as a predictor,
wherein:
the slice segment header of the picture P2 conditionally includes at least one syntax element inter_layer_pred_layer under the condition of the presence of explicit inter-layer prediction;
the syntax element inter_layer_pred_layer is indicative of the use of information from the picture P0 as a predictor, and
based on this indication, a reference to the picture P0 is included in a reference picture list when decoding the picture P2.
2. The method of claim 1 , wherein a flag inter_layer_pred_enabled_flag is set to 1 as a required condition for the presence of inter-layer prediction.
3. The method of claim 1 , wherein a presence of a syntax element inter_layer_pred_layer_idc indicates an explicit inter layer prediction relationship.
4. The method of claim 2 , wherein the flag inter_layer_pred_enabled_flag is a syntax element and part of the slice segment header.
5. The method of claim 1 , wherein the slice segment header comprises a syntax element num_inter_layer_ref_pic_minus1 indicative of the number of reference picture list entries that are used for inter-layer prediction relationships.
6. The method of claim 5 , further comprising:
using explicit reference layer prediction when a value of the syntax element num_inter_layer_ref_pic_minus1 is not equal to a number of reference pictures used by inter-layer prediction NumDirectRefLayers.
7. The method of claim 1 , wherein the presence of the syntax element inter_layer_pred_layer is conditioned on a flag max_one_active_ref_layer_flag located in a Video Parameter Set (VPS).
8. The method of claim 6 , wherein a number of inter_layer_pred_layer_idc syntax elements is NumActiveRefLayerPics, and NumActiveRefLayerPics is derived from a combination of at least two of the NumDirectRefLayers, all_ref_layers_active_flag, inter_layer_pred_enabled_flag and max_one_active_ref_layer_flag.
9. The method of claim 1 , wherein inter_layer_pred_layer_idc is coded in a fixed length binary code, wherein the length of the fixed length binary code derived from a NumberActiveRefLayerPics.
10. The method of claim 1 , further comprising reconstructing at least one treeblock of P2, wherein the treeblock of P2 is predicted at least in part from P0, and a corresponding prediction relationship is part of a treeblock header associated with the treeblock of P2.
11. The method of claim 10 further comprising reconstructing at least one treeblock of P2, wherein the treeblock of P2 is further predicted at least in part from P1 and a corresponding prediction relationship is part of a treeblock header associated with the treeblock of P2.
12. The method of claim 10 further comprising reconstructing at least one treeblock of P2, wherein the at least one treeblock of P2 is further predicted at least in part from P1 and no inter layer prediction information is present in the treeblock header associated with the treeblock of P2 that relates to a tree block of an inter layer prediction relationship between P2 and P1.
13. A system for decoding a bitstream comprising at least three layers or views, each including at least one picture P0, P1 and P2, respectively, wherein, P2 is predicted at least in part from P1, and P1 is predicted at least in part from P0 without any data indicating explicit inter-layer prediction data in the bitstream, wherein P2 is predicted at least in part from P0 in the presence of explicit inter-layer prediction,
a decoder (comprising a combination of hardware and software) configured to:
decode a slice segment header of the picture P2; and
reconstruct at least one sample of the picture P2 using information from the picture P0 as a predictor,
wherein:
the slice segment header of the picture P2 conditionally includes at least one syntax element inter_layer_pred_layer under the condition of the presence of explicit inter-layer prediction;
the syntax element inter_layer_pred_layer is indicative of the use of information from the picture P0 as a predictor, and
based on this indication, a reference to the picture P0 is included in a reference picture list when decoding the picture P2.
14. The system of claim 13 , wherein a flag inter_layer_pred_enabled_flag is set to 1 as a required condition for the presence of inter-layer prediction.
15. The system of claim 13 , wherein a presence of a syntax element inter_layer_pred_layer_idc indicates an explicit inter layer prediction relationship.
16. The system of claim 14 , wherein the flag inter_layer_pred_enabled_flag is a syntax element and part of the slice segment header.
17. The system of claim 13 , wherein the slice segment header comprises a syntax element num_inter_layer_ref_pic_minus1 indicative of the number of reference picture list entries that are used for inter-layer prediction relationships.
18. The system of claim 17 , wherein the decoder is further configured to:
use explicit reference layer prediction when a value of the syntax element num_inter_layer_ref_pic_minus1 is not equal to a number of reference pictures used by inter-layer prediction NumDirectRefLayers.
19. The system of claim 13 , wherein the presence of the syntax element inter_layer_pred_layer is conditioned on a flag max_one_active_ref_layer_flag located in a Video Parameter Set (VPS).
20. The system of claim 18 , wherein a number of inter_layer_pred_layer_idc syntax elements is NumActiveRefLayerPics, and NumActiveRefLayerPics is derived from a combination of at least two of the NumDirectRefLayers, all_ref_layers_active_flag, inter_layer_pred_enabled_flag and max_one_active_ref_layer_flag.
21. The system of claim 13 , wherein inter_layer_pred_layer_idc is coded in a fixed length binary code, wherein the length of the fixed length binary code derived from a NumberActiveRefLayerPics.
22. The system of claim 13 , wherein the decoder is further configured to: reconstruct at least one treeblock of P2, wherein the treeblock of P2 is predicted at least in part from P0, and a corresponding prediction relationship is part of a treeblock header associated with the treeblock of P2.
23. The system of claim 22 wherein the decoder is further configured to: reconstruct at least one treeblock of P2, wherein the treeblock of P2 is further predicted at least in part from P1 and a corresponding prediction relationship is part of a treeblock header associated with the treeblock of P2.
24. The system of claim 22 wherein the decoder is further configured to: reconstruct at least one treeblock of P2, wherein the at least one treeblock of P2 is further predicted at least in part from P1 and no inter layer prediction information is present in the treeblock header associated with the treeblock of P2 that relates to a tree block of an inter layer prediction relationship between P2 and P1.
25. A non-transitory computer readable medium comprising a set of exectuable instructions to direct a processor to perform the method of claim 1 .