Support of mixed IRAP and non-IRAP pictures within an access unit in multi-layer video bitstreams

1. A method of decoding implemented by a video decoder, comprising:

receiving a bitstream including a coded video sequence start (CVSS) access unit (AU), wherein the CVSS AU contains a picture unit (PU) for each layer, and wherein a coded picture in each PU is a coded layer video sequence start (CLVSS) picture;

identifying the coded picture from one of the layers based on a picture order count (POC) value; and

decoding the coded picture to obtain a decoded picture.

2. The method of claim 1 , wherein the CLVSS picture is an intra random access point (IRAP) picture.

3. The method of claim 1 , wherein the CLVSS picture is a gradual decoding refresh (GDR) picture.

4. The method of claim 1 , wherein the CLVSS picture is associated with a no output before recovery flag, and wherein the no output before recovery flag is equal to 1.

5. The method of claim 1 , wherein each layer of the CVSS AU is specified by a video parameter set (VPS).

6. The method of claim 1 , wherein the CVSS AU is an initial AU of a coded video sequence (CVS).

7. The method of claim 1 , further comprising displaying the coded picture as decoded on a display of an electronic device.

8. The method of claim 1 , further comprising:

receiving a second video bitstream including a second CVSS AU, wherein the second CVSS AU contains a picture unit (PU) for each layer, and wherein a coded picture in each PU is not a CLVSS picture; and

in response to the receiving, taking some other corrective measures to ensure that a conforming bitstream corresponding to the second video bitstream is received prior to decoding the coded picture.

9. A method of encoding implemented by a video encoder, the method comprising:

inserting a picture unit (PU) containing a coded picture in each layer of a coded video sequence start (CVSS) access unit (AU), wherein the video encoder is constrained such that the coded picture in each layer is a coded layer video sequence start (CLVSS) picture;

encoding a coded layer video sequence (CLVS) including the CVSS AU into a bitstream; and

storing the bitstream for communication toward a video decoder.

10. The method of claim 9 , wherein the CLVSS picture is an intra random access point (IRAP) picture.

11. The method of claim 9 , wherein the CLVSS picture is a gradual decoding refresh (GDR) picture.

12. The method of claim 9 , wherein each layer of the CVSS AU is specified by a video parameter set (VPS).

13. A decoding device, comprising:

a receiver configured to receive a bitstream including a coded video sequence start (CVSS) access unit (AU), wherein the CVSS AU contains a picture unit (PU) for each layer, and wherein a coded picture in each PU is a coded layer video sequence start (CLVSS) picture;

a memory coupled to the receiver, the memory storing instructions; and

one or more processors coupled to the memory, the one or more processors configured to execute the instructions to cause the decoding device to:

identify the coded picture from one of the layers based on a picture order count (POC) value; and

decode the coded picture to obtain a decoded picture.

14. The decoding device of claim 13 , wherein the CLVSS picture is an intra random access point (IRAP) picture.

15. The decoding device of claim 13 , wherein the CLVSS picture is a gradual decoding refresh (GDR) picture.

16. The decoding device of claim 13 , wherein each layer of the CVSS AU is specified by a video parameter set (VPS).