IP Library Granted Patent US 10,425,651
Granted Patent B2
US 10,425,651 · App. 16/157,723 · Granted Sep 24, 2019

Cluster-based dependency signaling

Inventors: Karsten Suehring (Berlin, DE); Thomas Schierl (Berlin, DE); Detlev Marpe (Berlin, DE); Robert Skupin (Berlin, DE); Yago Sanchez De La Fuente (Berlin, DE); Gerhard Tech (Berlin, DE)
Assignee: GE VIDEO COMPRESSION, LLC
H04N19/30H04L69/324H04N19/119H04N19/166H04N19/174H04N19/187H04N19/33H04N19/423H04N19/44H04N19/50H04N19/593H04N19/65H04N19/66H04N19/70H04N19/85H04N19/89H04N19/91
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,425,651
App. No.
16/157,723
Granted
Sep 24, 2019
Kind
B2
Abstract

The signalization of the inter-layer dependencies between layers of a multi-layered data stream is described. A good compromise between a too intensive restriction of the potential diversity of inter-layer dependencies on the one hand and a too complex signaling of the inter-layer dependencies on the other hand has been found by describing the inter-layer dependencies by way of a first inter-dependency syntax structure indicating inter-dependencies between pairs of different values representable by a base layer-ID and a second inter-dependency syntax structure indicating inter-dependencies between pairs of different values representable by an extension layer-ID, the base layer ID and extension layer ID indexing the layers the portions of the multi-layer data stream are associated with. In accordance with this concept, emphasis may be shifted between increased diversity of the signalizable inter-layer dependencies on the one hand and reduced side-information overhead for signaling the inter-layer dependencies on the other hand.

Claims (59)

1. A device configured to process a data stream representing a multi-layered signal into which a video is coded using inter-layer prediction as different layers, which correspond to different levels of information amount, the data stream comprising a plurality of packets, each of which is associated with one of the different layers, each layer being indexed by a base layer-ID or a base layer-ID and an extension layer-ID, the device configured to:

read, from the data stream,

a first inter-dependency syntax structure indicating inter-dependencies between pairs of values representable by the base layer-ID, and

a second inter-dependency syntax structure indicating inter-dependencies between pairs of values representable by the extension layer-ID, wherein an instantiation of the second inter-dependency syntax structure indicates (a) a first layer dependent on, in terms of inter-layer prediction, a second layer, wherein the first and second layers are associated with a same base layer-ID, or (b) a first layer dependent on, in terms of inter-layer prediction, a second layer, wherein the first and second layers are associated with different base layer-IDs and different extension layer-IDs; and

construct, based on the first and second inter-dependency syntax structures, an inter-layer dependency description revealing possible inter-layer prediction dependencies between the different layers.

2. The device according to claim 1 , wherein the device is configured to:

read a plurality of instantiations of the second inter-dependency structure from the data stream, namely for different pairs (j,k) of values 0<k≤j≤n<N representable by the base layer-ID, with N being the number of values representable by the base layer-ID, by

reading, or suppressing reading, the second inter-dependency structure from the data stream for pair (j,k) conditionally depending on the first inter-dependency syntax structure indicating a presence or non-presence of interdependency between pair (j,k),

wherein one of the plurality of instantiations of the second inter-dependency structure read for a pair (j,k) with j=k, indicates which layers among those comprising base layer-ID j are inter-layer prediction dependent on other layers among those comprising base layer-ID j, and another of the plurality of instantiations of the second inter-dependency structure read for pair (j,k), with k<j indicates which layers among those comprising base layer-ID j are inter-layer prediction dependent on layers among those comprising base layer-ID k.

3. The device according to claim 1 , wherein the device is configured to:

read instantiations of the second inter-dependency structure from the data stream, along with, for each instantiation of the second inter-dependency structure, an association syntax structure which associates the respective instantiation of the second inter-dependency structure with a respective subset of pairs (j,k) of values 0<k≤j≤n<N representable by the base layer-ID, with N being the number of values representable by the base layer-ID.

4. The device according to claim 3 , wherein the association syntax structure comprises a flag for each pair (j,k) and the device is configured to:

parse the association syntax structure with skipping pairs (j,k) of values 0<k≤j≤n<N for which the first inter-dependency syntax structure indicates the non-presence of interdependency between pair (j,k).

5. The device according to claim 1 , wherein the number of values representable by the extension layer-ID is M and the second inter-dependency syntax structure comprises

one binary value for each pair (p,q) of extension values 0<q<p≤m<M, in order to indicate which layers among those comprising base layer-ID j are inter-layer prediction dependent on other layers among those comprising base layer-ID j, and

one binary value for each pair (p,q) of extension values 0<p,q≤m<M, in order to indicate which layer with extension value p and comprising base layer-ID j is inter-layer prediction dependent on layer with extension value q and comprising base layer-ID k<j.

6. The device according to claim 1 , wherein the device is configured to derive n from an explicitly signaling of a maximally assumed value among the N values representable by the base layer-ID, in the data stream, and/or derive m from an explicitly signaling of a maximally assumed value among the M values representable by the extension layer-ID, in the data stream.

7. The device according to claim 1 , where the first inter-dependency syntax structure defines interdependencies between different sets of layers, each of the different sets being associated with a unique base layer-ID.

8. The device according to claim 1 , wherein the device is configured to discard packets in units of sets of layers each comprising a common base layer-ID based on the first inter-dependency syntax structure and the packets' base layer-ID.

9. The device according to claim 8 , wherein the device is configured to discard further individual layers based on the first and second inter-dependency syntax structure and the packets' base layer-ID and the packets' extension layer-ID.

10. The device according to claim 1 , wherein the device is

a video decoder configured to decode the multi-layered video signal using the inter-layer prediction and the inter-layer dependency matrix.

11. The device according to claim 1 , wherein the multi-layered video signal has the video coded thereinto at the different layers using inter-layer prediction such that any layer inter-layer predicted from another layer adds one of:

further views;

depth information;

alpha blending information;

color component information;

spatial resolution refinement; and

SNR resolution refinement.

12. A method for processing a data stream representing a multi-layered signal into which a video is coded using inter-layer prediction as different layers, which correspond to different levels of information amount, data stream comprising a plurality of packets, each of which is associated with one of the different layers, each layer being indexed by a base layer-ID or a base layer-ID and an extension layer-ID, the method comprising:

reading, from the data stream,

a first inter-dependency syntax structure indicating interdependencies between pairs of values representable by the base layer-ID, and

a second inter-dependency syntax structure indicating inter-dependencies between pairs of values representable by the extension layer-ID, wherein an instantiation of the second inter-dependency syntax structure indicates (a) a first layer dependent on, in terms of inter-layer prediction, a second layer, wherein the first and second layers are associated with a same base layer-ID, or (b) a first layer dependent on, in terms of inter-layer prediction, a second layer, wherein the first and second layers are associated with different base layer-IDs and different extension layer-IDs; and

constructing, based on the first and second inter-dependency syntax structures, an inter-layer dependency description revealing possible inter-layer prediction dependencies between the different layers.

13. The method according to claim 12 , further comprising:

reading a plurality of instantiations of the second inter-dependency structure from the data stream, namely for different pairs (j,k) of values 0<k≤j≤n<N representable by the base layer-ID, with N being the number of values representable by the base layer-ID, by

reading, or suppressing reading, the second inter-dependency structure from the data stream for pair (j,k) conditionally depending on the first inter-dependency syntax structure indicating a presence or non-presence of interdependency between pair (j,k),

wherein one of the plurality of instantiations of the second inter-dependency structure read for a pair (j,k) with j=k, indicates which layers among those comprising base layer-ID j are inter-layer prediction dependent on other layers among those comprising base layer-ID j, and another of the plurality of instantiations of the second inter-dependency structure read for pair (j,k), with k<j indicates which layers among those comprising base layer-ID j are inter-layer prediction dependent on layers among those comprising base layer-ID k.

14. The method according to claim 12 , further comprising:

reading instantiations of the second inter-dependency structure from the data stream, along with, for each instantiation of the second inter-dependency structure, an association syntax structure which associates the respective instantiation of the second inter-dependency structure with a respective subset of pairs (j,k) of values 0<k≤j≤n<N representable by the base layer-ID, with N being the number of values representable by the base layer-ID.

15. The method according to claim 14 , wherein the association syntax structure comprises a flag for each pair (j,k) and the method further comprises:

parsing the association syntax structure with skipping pairs (j,k) of values 0<k≤j≤n<N for which the first inter-dependency syntax structure indicates the non-presence of interdependency between pair (j,k).

16. The method according to claim 12 , wherein the number of values representable by the extension layer-ID is M and the second inter-dependency syntax structure comprises

one binary value for each pair (p,q) of extension values 0<q<p≤m<M, in order to indicate which layers among those comprising base layer-ID j are inter-layer prediction dependent on other layers among those comprising base layer-ID j, and

one binary value for each pair (p,q) of extension values 0<p,q≤m<M, in order to indicate which layer with extension value p and comprising base layer-ID j is inter-layer prediction dependent on layer with extension value q and comprising base layer-ID k<j.

17. The method according to claim 12 , further comprising discarding packets in units of sets of layers each comprising a common base layer-ID based on the first inter-dependency syntax structure and the packets' base layer-ID.

18. The method according to claim 17 , further comprising discarding further individual layers based on the first and second inter-dependency syntax structure and the packets' base layer-ID and the packets' extension layer-ID.

19. An encoder configured to encode into a data stream representing a multi-layered signal into which a video using inter-layer prediction as different layers, which correspond to different levels of information amount, the data stream comprising a plurality of packets, each of which is associated with one of the different layers, each layer being indexed by a base layer-ID or a base layer-ID and an extension layer-ID, the encoder configured to:

insert, into the data stream,

a first inter-dependency syntax structure indicating interdependencies between pairs of values representable by the base layer-ID, and

a second inter-dependency syntax structure indicating inter-dependencies between pairs of values representable by the extension layer-ID, wherein an instantiation of the second inter-dependency syntax structure indicates (a) a first layer dependent on, in terms of inter-layer prediction, a second layer, wherein the first and second layers are associated with a same base layer-ID, or (b) a first layer dependent on, in terms of inter-layer prediction, a second layer, wherein the first and second layers are associated with different base layer-IDs and different extension layer-IDs,

with setting the first and second inter-dependency syntax structures such that, based on the first and second inter-dependency syntax structures, an inter-layer dependency description revealing possible inter-layer prediction dependencies between the different layers is constructible.

20. The encoder according to claim 19 , wherein the multi-layered video signal has the video coded thereinto at the different layers using inter-layer prediction such that any layer inter-layer predicted from another layer adds one of:

further views;

depth information;

alpha blending information;

color component information;

spatial resolution refinement; and

SNR resolution refinement.

Assignments (3)
CHANGE OF NAME Recorded Jan 30, 2026
From: GE VIDEO COMPRESSION, LLC
To: DOLBY VIDEO COMPRESSION, LLC
Reel/Frame 074536/0717 →
CHANGE OF NAME Recorded Nov 26, 2024
From: GE VIDEO COMPRESSION, LLC
To: DOLBY VIDEO COMPRESSION, LLC
Reel/Frame 069450/0425 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 11, 2018
From: SUEHRING, KARSTEN; SCHIERL, THOMAS; MARPE, DETLEV; SKUPIN, ROBERT; SANCHEZ DE LA FUENTE, YAGO; TECH, GERHARD
To: GE VIDEO COMPRESSION, LLC
Reel/Frame 047216/0367 →
Continuity (4)
Continuation 14995222 · Jan 14, 2016
Continuation PCTEP2014065181 · Jul 15, 2014
Provisional Application 61846479 · Jul 15, 2013
Related Publication 20190082187A1 · Mar 14, 2019
Cited By (3)
US 12,294,727 US 12,593,054 US 12,621,473