System and method for automatically generating split screen for a video of a dynamic scene

1. One or more non-transitory computer readable storage mediums storing one or more sequences of instructions, which when executed by one or more processors, causes automatic generation of split screen for a video of a dynamic scene, by

obtaining the video of the dynamic scene that is captured by a static camera;

detecting one or more objects in the video of the dynamic scene using an object detection technique;

selecting one or more shot specifications for the one or more objects in the video based on inputs of a user;

automatically cropping at least one of (a) the dynamic scene of the one or more objects or (b) the dynamic scene for the possible combinations of the one or more objects in the video based on the selected shot specifications;

automatically selecting a layout configuration for the split screen based on the cropped dynamic scenes of the one or more objects to partition a display screen of a computing unit into (a) a top half and (b) a bottom half with one or more partition views;

automatically optimizing, using a dynamic programming optimization technique, the split screen by (a) seamlessly varying the one or more partition views, (b) providing smoothness in the cropped dynamic scenes irrespective of the camera motion and (c) providing jerk free transitions when the one or more objects move from one partition view to another partition view;

automatically generating the split screen by depicting (a) an overview of the original shot of the video in the top half and (b) the cropped dynamic scenes in the one or more partition views of the bottom half; and

displaying the split screen in real time onto a display of the computing unit.

2. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein the optimization of the one or partition views is performed based on various factors that comprise at least one of (a) the one or more shot specifications, (b) inclusion of desired objects or (c) a steady camera movement.

3. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein the video of the dynamic scene is captured by at least one of (a) a single static camera or (b) multiple static cameras positioned at a vantage point covering the entire dynamic scene.

4. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein the layout configuration is automatically selected based on at least one of (i) a proximity between the one or more objects in the video and (ii) the one or more objects in the video.

5. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein the one or more shot specifications is selected from a group comprising (a) a medium close up shot specification, (b) a medium shot specification or (c) a full shot specification.

6. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein the optimization of the split screen is performed using a top down layout constraints technique, wherein the top down layout constraints technique (a) provides the jerk free transitions and (b) seamlessly varies the one or more partition views by at least one of (i) concatenation of one or more subset shot specifications equal to a higher order shot specification or (ii) splitting of the higher order shot specification equal to the one or more subset shot specifications.

7. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein the selected shot specification provides finer details of (a) emotions, (b) lip movements and (c) body languages of the one or more objects in the video.

8. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 1 , wherein further analyzes a subtitle in the video and provides the appropriate subtitle for the object in the associated partition view.

9. The one or more non-transitory computer readable storage mediums storing one or more sequences of instructions of claim 6 , wherein concatenates the one or more subset shot specifications equal to the higher order shot specification when the proximity between the one or more objects in the video becomes less than a threshold proximity, wherein further automatically splits the higher order shot specification equal to the one or more subset shot specifications when the proximity between the one or more objects in the video becomes more than the threshold proximity.

10. A system for automatically generating split screen for a video of a dynamic scene, comprising:

a memory that stores (a) a set of instructions and (b) a database that stores an original shot of the dynamic scene, dimensions of the split screen of the video of the dynamic scene, a threshold proximity and one or more layout configurations;

characterized in that a specialized split screen generation processor that is configured to

obtain the video of the dynamic scene that is captured by a static camera;

detect one or more objects in the video of the dynamic scene using an object detection technique;

select one or more shot specifications for the one or more objects in the video based on inputs of a user;

automatically crops at least one of (a) the dynamic scene of the one or more objects or (b) the dynamic scenes for the possible combinations of the one or more objects in the video based on the selected shot specifications;

automatically select a layout configuration for the split screen based on the cropped dynamic scenes of the one or more objects to partition a display screen of a computing unit into (a) a top half and (b) a bottom half with one or more partition views;

automatically optimize, using a dynamic programming optimization technique, the split screen by (a) seamlessly varying the one or more partition views, (b) providing smoothness in the cropped dynamic scenes irrespective of the camera motion and (c) providing jerk free transitions when the one or more objects move from one partition view to another partition view;

automatically generate the split screen by depicting (a) an overview of the original shot of the video in the top half and (b) the cropped dynamic scenes in the one or more partition views of the bottom half; and

automatically display the split screen in real time onto a display of the computing unit.

11. The system of claim 10 , wherein the optimization of the one or partition views is performed based on various factors that comprise at least one of (a) the one or more shot specifications, (b) inclusion of desired objects or (c) a steady camera movement.

12. The system of claim 10 , wherein the video of the dynamic scene is captured by at least one of (a) a single static camera or (b) multiple static cameras positioned at a vantage point covering the entire dynamic scene.

13. The system of claim 10 , wherein the layout configuration is automatically selected based on at least one of (i) a proximity between the one or more objects in the video and (ii) the one or more objects in the video.

14. The system of claim 10 , wherein the one or more shot specifications is selected from a group comprising (a) a medium close up shot specification, (b) a medium shot specification or (c) a full shot specification.

15. The system of claim 10 , wherein the optimization of the split screen is performed using a top down layout constraints technique, wherein the top down layout constraints technique (a) provides the jerk free transitions and (b) seamlessly varies the one or more partition views by at least one of (i) concatenation of one or more subset shot specifications equal to a higher order shot specification or (ii) splitting of the higher order shot specification equal to the one or more subset shot specifications.

16. The system of claim 10 , wherein the selected shot specification provides finer details of (a) emotions, (b) lip movements and (c) body languages of the one or more objects in the video.

17. The system of claim 10 , wherein the system analyzes a subtitle in the video and provides the appropriate subtitle for the object in the associated partition view.

18. The system of claim 15 , wherein the system automatically concatenates the one or more subset shot specifications equal to the higher order shot specification when the proximity between the one or more objects in the video becomes less than the threshold proximity, wherein the system automatically splits the higher order shot specification equal to the one or more subset shot specifications when the proximity between the one or more objects in the video becomes more than the threshold proximity.

19. A computer implemented method for automatically generating split screen for a video of a dynamic scene, comprising:

obtaining the video of the dynamic scene that is captured by a static camera;

automatically detecting one or more objects in the video of the dynamic scene using an object detection technique;

selecting one or more shot specifications for the one or more objects in the video based on inputs of a user;

automatically cropping at least one of (a) the dynamic scene of the one or more objects or (b) the dynamic scenes of the possible combinations for the one or more objects in the video based on the selected shot specifications;

automatically selecting a layout configuration for the split screen based on the cropped dynamic scenes to partition a display screen of a computing unit into (a) a top half and (b) a bottom half with one or more partition views;

automatically optimizing, using a dynamic programming optimization technique, the split screen by (a) seamlessly varying the one or more partition views, (b) providing smoothness in the cropped dynamic scenes irrespective of the camera motion and (c) providing jerk free transitions when the one or more objects move from one partition view to another partition view;

automatically generating the split screen by depicting (a) an overview of an original shot of the video in the top half and (b) the cropped dynamic scenes in the one or more partition views of the bottom half; and

displaying the split screen in real time onto a display of the computing unit.