Inter prediction for predictive geometry coding

1 . A method of processing point cloud data, the method comprising:

determining that resampling is to be applied to a first reference frame for a first slice of the point cloud data or a first frame of the point cloud data;

applying resampling to the first reference frame to generate a resampled reference frame;

determining one or more first inter prediction candidates based on the resampled reference frame;

processing the first slice of the point cloud data or the first frame of the point cloud data based on the one or more first inter prediction candidates;

determining that global motion compensation is not to be applied to a second slice of the point cloud data or a second frame of the point cloud data;

based on the determination that global motion compensation is not to be applied to the second slice of the point cloud data or the second frame of the point cloud data, refraining from including any inter prediction candidates from any global motion compensated frames in an inter predictor candidate list; and

processing the second slice of the point cloud data or the second frame of the point cloud data based on the inter predictor candidate list.

2 . The method of claim 1 , further comprising:

signaling or parsing a flag in a bitstream, the flag being indicative of whether resampling is to be applied to the first reference frame.

3 . The method of claim 2 , wherein the flag comprises a syntax element indicative of whether resampling is to be applied to the first reference frame of the first slice of the point cloud data or the first frame of the point cloud data and wherein the syntax element is signaled in a parameter set or a slice header.

4 . The method of claim 3 , wherein the syntax element is signaled in a geometry parameter set (GPS).

5 . The method of claim 3 , wherein the syntax element is signaled in the slice header and wherein values of corresponding syntax elements for corresponding slices of a same frame as the first slice of the point cloud data have a same value.

6 . The method of claim 3 , wherein the slice header comprises a data unit header and the syntax element is signaled in the data unit header.

7 . The method of claim 1 , further comprising determining one or more additional inter prediction candidates based on at least one of an uncompensated reference frame or a global-motion-compensated reference frame, wherein processing the first slice of the point cloud data or the first frame of the point cloud data is further based on the one or more additional inter prediction candidates.

8 . The method of claim 7 , wherein the uncompensated reference frame comprises the first reference frame.

9 . The method of claim 7 , further comprising generating the global-motion-compensated reference frame based on the first reference frame.

10 . The method of claim 1 , further comprising applying resampling to all inter prediction candidates of the first reference frame for the first slice of the point cloud data or the first frame of the point cloud data.

11 . The method of claim 1 , further comprising:

determining not to apply resampling to a third reference frame;

applying global motion compensation to the third reference frame to generate a global-motion-compensated reference frame;

determining, based on the determination not to apply resampling to the third reference frame, one or more third inter prediction candidates based on the global-motion-compensated reference frame; and

processing a third slice of the point cloud data or a third frame of the point cloud data based on the one or more third inter prediction candidates.

12 . The method of claim 11 , wherein the one or more third inter prediction candidates comprise all third inter prediction candidates for the third reference frame.

13 . The method of claim 1 , wherein the one or more first inter prediction candidates based on the resampled reference frame comprise one or more first resampled inter prediction candidates, the method further comprising determining one or more additional first inter prediction candidates based on a global-motion-compensated reference frame, wherein processing the first slice of the point cloud data or the first frame of the point cloud data is further based on the one or more additional first inter prediction candidates.

14 . The method of claim 13 , further comprising:

determining not to apply resampling to a third reference frame;

determining one or more third inter prediction candidates based on an uncompensated third reference frame;

applying global motion compensation to the third reference frame to generate a global-motion-compensated reference frame;

determining one or more additional third inter prediction candidates based on the global-motion-compensated reference frame and the determination not to apply resampling to the third reference frame; and

processing a third slice of the point cloud data or a third frame of the point cloud data based on the one or more third inter prediction candidates and the one or more additional third inter prediction candidates.

15 . The method of claim 1 , wherein the resampled reference frame is a first resampled reference frame, the method further comprising determining one or more second inter predictor candidates for the inter predictor candidate list based on at least one of an unsampled reference frame or a second resampled reference frame.

16 . The method of claim 15 , further comprising:

signaling or parsing a flag in a bitstream indicative of whether resampling is to be applied to a second reference frame;

determining to apply resampling to the second reference frame; and

applying resampling to the second reference frame to generate the second resampled reference frame based on the determination to apply resampling to the second reference frame.

17 . The method of claim 1 , wherein the determination that global motion compensation is not to be applied to the second slice of the point cloud data or the second frame of the point cloud data is based on at least one of:

a flag signaled in a bitstream indicative of global motion compensation not being applied;

global motion parameters associated with the second slice of the point cloud data or the second frame of the point cloud data not being signaled in the bitstream; or

global motion parameters associated with the second slice of the point cloud data or the second frame of the point cloud data corresponding to zero rotation and zero translation.

18 . The method of claim 1 , further comprising generating the point cloud.

19 . A device for processing a point cloud, the device comprising:

one or more memories configured to store point cloud data; and

one or more processors implemented in circuitry and communicatively coupled to the one or more memories, the one or more processors being configured to:

determine that resampling is to be applied to a first reference frame for a first slice of the point cloud data or a first frame of the point cloud data;

apply resampling to the first reference frame to generate a resampled reference frame;

determine one or more first inter prediction candidates based on the resampled reference frame;

process the first slice of the point cloud data or the first frame of the point cloud data based on the one or more first inter prediction candidates;

determine that global motion compensation is not to be applied to a second slice of the point cloud data or a second frame of the point cloud data;

based on the determination that global motion compensation is not to be applied to the second slice of the point cloud data or the second frame of the point cloud data, refrain from including any inter prediction candidates from any global motion compensated frames in an inter predictor candidate list; and

process the second slice of the point cloud data or the second frame of the point cloud data based on the inter predictor candidate list.

20 . The device of claim 19 , wherein the one or more processors are further configured to:

signal or parse a flag in a bitstream, the flag being indicative of whether resampling is to be applied to the first reference frame.

21 . The device of claim 20 , wherein the flag comprises a syntax element indicative of whether resampling is to be applied to the first reference frame of the first slice of the point cloud data or the first frame of the point cloud data and wherein the syntax element is signaled in a parameter set or a slice header.

22 . The device of claim 21 , wherein the syntax element is signaled in a geometry parameter set (GPS).

23 . The device of claim 21 , wherein the syntax element is signaled in the slice header and wherein values of corresponding syntax elements for corresponding slices of a same frame as the first slice of the point cloud data have a same value.

24 . The device of claim 21 , wherein the slice header comprises a data unit header and the syntax element is signaled in the data unit header.

25 . The device of claim 19 , wherein the one or more processors are further configured to determine one or more additional inter prediction candidates based on at least one of an uncompensated reference frame or a global-motion-compensated reference frame, wherein the one or more processors are configured to process the first slice of the point cloud data or the first frame of the point cloud data further based on the one or more additional inter prediction candidates.

26 . The device of claim 25 , wherein the uncompensated reference frame comprises the first reference frame.

27 . The device of claim 25 , wherein the one or more processors are further configured to generate the global-motion-compensated reference frame based on the first reference frame.

28 . The device of claim 19 , further comprising applying resampling to all inter prediction candidates of the first reference frame for the first slice of the point cloud data or the first frame of the point cloud data.

29 . The device of claim 19 , wherein the one or more processors are further configured to:

determine not to apply resampling to a third reference frame;

apply global motion compensation to the third reference frame to generate a global-motion-compensated reference frame;

determine, based on the determination not to apply resampling to the second third reference frame, one or more second third inter prediction candidates based on the global-motion-compensated reference frame; and

process a third slice of the point cloud data or a third frame of the point cloud data based on the one or more third inter prediction candidates.

30 . The device of claim 29 , wherein the one or more third inter prediction candidates comprise all third inter prediction candidates for the third reference frame.

31 . The device of claim 19 , wherein the one or more first inter prediction candidates based on the resampled reference frame comprise one or more first resampled inter prediction candidates, and wherein the one or more processors are further configured to determine one or more additional first inter prediction candidates based on a global-motion-compensated reference frame, wherein the one or more processors process the first slice of the point cloud data or the first frame of the point cloud data further based on the one or more additional first inter prediction candidates.

32 . The device of claim 31 , wherein the one or more processors are further configured to:

determine not to apply resampling to a third reference frame;

determine one or more third inter prediction candidates based on an uncompensated third reference frame;

apply global motion compensation to the third reference frame to generate a global-motion-compensated reference frame;

determine one or more additional third inter prediction candidates based on the global-motion-compensated reference frame and the determination not to apply resampling to the third reference frame; and

process a third slice of the point cloud data or a third frame of the point cloud data based on the one or more third inter prediction candidates and the one or more additional third inter prediction candidates.

33 . The device of claim 19 , wherein the resampled reference frame is a first resampled reference frame, and wherein the one or more processors are further configured to determine one or more second inter predictor candidates for the inter predictor candidate list based on at least one of an unsampled reference frame or a second resampled reference frame.

34 . The device of claim 33 , wherein the one or more processors are further configured to:

signal or parse a flag in a bitstream indicative of whether resampling is to be applied to a second reference frame;

determine to apply resampling to the second reference frame; and

apply resampling to the second reference frame to generate the second resampled reference frame based on the determination to apply resampling to the second reference frame.

35 . The device of claim 19 , wherein the determination that global motion compensation is not to be applied to the second slice of the point cloud data or the second frame of the point cloud data is based on at least one of:

a flag signaled in a bitstream indicative of global motion compensation not being applied;

global motion parameters associated with the second slice of the point cloud data or the second frame of the point cloud data not being signaled in the bitstream; or

global motion parameters associated with the second slice of the point cloud data or the second frame of the point cloud data corresponding to zero rotation and zero translation.

36 . The device of claim 19 , wherein the one or more processors are further configured to generate the point cloud.

37 . The device of claim 19 , further comprising a display configured to present imagery based on the point cloud.

38 . Non-transitory computer-readable storage media having stored thereon instructions that, when executed, cause one or more processors to:

determine that resampling is to be applied to a first reference frame for a first slice of point cloud data or a first frame of the point cloud data;

apply resampling to the first reference frame to generate a resampled reference frame;

determine one or more first inter prediction candidates based on the resampled reference frame;

process the first slice of the point cloud data or the first frame of the point cloud data based on the one or more first inter prediction candidates;

determine that global motion compensation is not to be applied to a second slice of the point cloud data or a second frame of the point cloud data;

based on the determination that global motion compensation is not to be applied to the second slice of the point cloud data or the second frame of the point cloud data, refrain from including any inter prediction candidates from any global motion compensated frames in an inter predictor candidate list; and

process the second slice of the point cloud data or the second frame of the point cloud data based on the inter predictor candidate list.

39 . A device for processing point cloud data, the device comprising:

means for determining that resampling is to be applied to a first reference frame for a first slice of the point cloud data or a first frame of the point cloud data;

means for applying resampling to the first reference frame to generate a resampled reference frame;

means for determining one or more first inter prediction candidates based on the resampled reference frame; and

means for processing the first slice of the point cloud data or the first frame of the point cloud data based on the one or more first inter prediction candidates;

means for determining that global motion compensation is not to be applied to a second slice of the point cloud data or a second frame of the point cloud data;

means for, based on the determination that global motion compensation is not to be applied to the second slice of the point cloud data or the second frame of the point cloud data, refraining from including any inter prediction candidates from any global motion compensated frames in an inter predictor candidate list; and

means for processing the second slice of the point cloud data or the second frame of the point cloud data based on the inter predictor candidate list.