System and signalling of video splitter and merger for parallel network based media processing
In accordance with example embodiments of the invention there is at least one method and apparatus to perform assigning, by a network device, a step descriptor to at least one segment split of a workflow comprising at least one data input instance from at least one network node; and sending at least one sub-flow of the workflow based on the at least one segment split towards at least one queued task for communication processing, wherein the communication processing is based at least on the step descriptor assigned to the at least one segment split.
1. A method, comprising:
assigning, by a network device, a step descriptor to at least one segment split of a workflow comprising at least one data input instance from at least one network node,
wherein the step descriptor comprises a sequence number and is using one metadata for each input for receiving the sequence number and comprises a start and a segment duration of each of the at least one data input instance, and
wherein for the at least one segment split comprising frame splitting, the step descriptor includes a same grid identifier timestamp and a same grid identification such that there can be multiple grid tasks for a same time stamp; and
sending at least one sub-flow of the workflow based on the at least one segment split comprising the at least one data input instance towards at least one queued task for communication processing, wherein the communication processing is based at least on the step descriptor assigned to the at least one segment split.
2. The method according to claim 1 , wherein the assigning comprises the step descriptor is assigned to a payload of at least one segment split passed through the at least one sub-flow, and wherein segment data of the workflow is encapsulated into the payload and passed through the at least one sub-flow without any modification.
3. The method according to claim 1 , wherein the at least one segment split is based on at least one of temporal segmentation or spatial segmentation, wherein the temporal segmentation uses a one-dimensional segment and the spatial segmentation uses a two-dimensional segment.
4. The method according to claim 1 , wherein the at least one segment split of the work flow is using correlation operations based on at least one of spatial segmentation or volumetric segmentation, wherein the volumetric segmentation uses a three-dimensional segment.
5. The method according to claim 1 , wherein the at least one segment split of the work flow is using correlation operations based on at least one of spatial or volumetric segmentation, or higher-degree segmentation, wherein the higher-degree segmentation uses segments determined by the number of dimensions and its divisor types.
6. The method according to claim 1 , wherein for a case the at least one segment strip is based on temporal segmentation, the step descriptor comprises at least one segment identification, a monotonically increased time stamp, and wherein the at least one segment identification comprises a monotonically increased timestamp for temporal segmentation.
7. The method according to claim 6 , wherein the monotonically increased timestamp is identical for each segment associated with the at least one segment identification.
8. The method according to claim 1 , wherein for a case the at least one segment strip is based on spatial segmentation, the step descriptor comprises at least one of segment identification, a timestamp, at least one of a regular or an irregular grid definition, or at least one of a regular or an irregular grid identifier.
9. The method according to claim 8 , wherein the segment identification together with at least one of a regular or an irregular grid identifier are unique to enable at least one of reassembly or merger of the at least one segment split of a workflow.
10. The method according to claim 9 , wherein the at least one of reassembly or merger takes into account grid tasks for at least one of a given temporal segment or a same timestamp value.
11. The method according to claim 1 , wherein the communication processing comprises at least one of reassembly or merger of the at least one segment split of the workflow.
12. The method according to claim 1 , wherein the communication processing comprises at least one of stateful or stateless execution of the at least one segment split of the workflow at the at least one queued task.
13. The method according to claim 1 , wherein the at least one segment split is dynamic, comprising the at least one segment split is based on a content of media data of the workflow.
14. The method according to claim 13 , wherein the at least one segment split is based on a video analysis task associated with the content of media data of the workflow.
15. The method according to claim 14 , wherein the assigning the step descriptor comprises assigning step descriptors with a particular region definition for the communication processing, wherein for a case of two dimensional segmentation and a high dimension segment divisor is two and dimension description is the particular region width and height.
16. The method according to claim 15 , wherein the particular region definition comprises a different or irregular region definition for storing the at least one segment split of a workflow for later communication processing.
17. An apparatus comprising:
at least one processor; and
at least one non-transitory memory including computer program code, where the at least one non-transitory memory and the computer program code are configured, with the at least one processor, to cause the apparatus to at least:
assign, by a network device, a step descriptor to at least one segment split of a workflow comprising at least one data input instance from at least one network node,
wherein the step descriptor comprises a sequence number and is using one metadata for each input for receiving the sequence number and at least one of comprises a start or and a segment duration of each of the at least one data input instance, and
wherein for the at least one segment split comprising frame splitting, the step descriptor includes a same grid identifier timestamp and a same grid identification such that there can be multiple grid tasks for a same time stamp; and
send at least one sub-flow of the workflow based on the at least one segment split comprising the at least one data input instance towards at least one queued task for communication processing, wherein the communication processing is based at least on the step descriptor assigned to the at least one segment split.
18. The apparatus according to claim 17 , wherein the assigning comprises the step descriptor is assigned to a payload of at least one segment split passed through the at least one sub-flow, and wherein segment data of the workflow is encapsulated into the payload and passed through the at least one sub-flow without any modification.
19. The apparatus according to claim 17 , wherein the at least one segment split is based on at least one of temporal segmentation or spatial segmentation, wherein the temporal segmentation uses a one-dimensional segment and the spatial segmentation uses a two-dimensional segment.
20. The apparatus according to claim 17 , wherein the at least one segment split of the work flow is using correlation operations based on at least one of spatial segmentation or volumetric segmentation, wherein the volumetric segmentation uses a three dimensional segment.
21. The apparatus according to claim 17 , wherein the at least one segment split of the work flow is using correlation operations based on at least one of spatial or volumetric segmentation, or higher-degree segmentation, wherein the higher-degree segmentation uses segments determined by the number of dimensions and its divisor types.
22. The apparatus according to claim 17 , wherein for a case the at least one segment strip is based on temporal segmentation, the step descriptor comprises at least one segment identification, a global timestamp, and wherein the at least one segment identification comprises a monotonically increased timestamp for temporal segmentation.
23. The apparatus according to claim 22 , wherein the monotonically increased timestamp is identical for each segment associated with the at least one segment identification.
24. The apparatus according to claim 17 , wherein for a case the at least one segment strip is based on spatial segmentation, the step descriptor comprises at least one of segment identification, a timestamp, at least one of a regular or an irregular grid definition, or at least one of a regular or an irregular grid identifier.
25. The apparatus according to claim 24 , wherein the segment identification together with at least one of a regular or an irregular grid identifier are unique to enable at least one of reassembly or merger of the at least one segment split of a workflow.
26. The apparatus according to claim 25 , wherein the at least one of reassembly or merger takes into account grid tasks for at least one of a given temporal segment or a same timestamp value.
27. The apparatus according to claim 24 , wherein the communication processing comprises at least one of reassembly or merger of the at least one segment split of the workflow.
28. The apparatus according to claim 17 , wherein the communication processing comprises at least one of stateful or stateless execution of the at least one segment split of the workflow at the at least one queued task.
29. The apparatus according to claim 28 , wherein the at least one segment split is based on a video analysis task associated with the content of media data of the workflow.
30. The apparatus according to claim 17 , wherein the at least one segment split is dynamic, comprising the at least one segment split is based on a content of media data of the workflow.
31. The apparatus according to claim 17 , wherein the assigning the step descriptor comprises assigning step descriptors with a particular region definition for the communication processing, wherein for a case of two dimensional segmentation and a high dimension segment divisor is two and dimension description is the particular region width and height.
32. The apparatus according to claim 31 , wherein the particular region definition comprises a different or irregular region definition for storing the at least one segment split of a workflow for later communication processing.