Live Streaming Startup Kiswe Pushes Multiple Camera Angles

Keep an eye on New Jersey-based startup Kiswe Mobile, which has developed a technology platform that could allow sports programmers, teams and leagues to market new live streaming video services that give viewers the ability to navigate multiple camera angles.

Jeong Kim
Kiswe founder and inventor Jeong Kim

Based in New Providence, about 10 miles from Verizon’s Basking Ridge headquarters, Kiswe’s leadership team includes veterans of Bell Labs and Alcatel-Lucent. Chairman and Co-founder Jeong Kim is named as lead inventor on a Kiswe patent application that The Donohue Report obtained on Thursday.

Kiswe has also developed an interactive live streaming app for iOS and Android devices called Trekker Live.

Abstract: A method comprises receiving content from at least one source, processing the content to identify an object in the content, assigning an identifier to the object based on the processing, permitting viewer selection of the object identifier, and providing additional information for presentation to a viewer of the content responsive to viewer selection of the object identifier. The content in some embodiments comprises live video from at least one live video source. For example, the content may comprise at least a portion of a panoramic video comprising a combination of multiple live videos from respective ones of a plurality of video sources. The panoramic video may be digitally cropped and scaled before a resulting version of the content is provided for presentation to the viewer. A plurality of different digitally cropped and scaled versions of the content from the panoramic video can be generated and made available to the viewer.

Patent Application

Claims:

1. A method comprising: receiving content from at least one source; processing the content to identify an object in the content; assigning an identifier to the object based on the processing; permitting viewer selection of the object identifier; and providing additional information for presentation to a viewer of the content responsive to viewer selection of the object identifier; wherein the receiving, processing, assigning, permitting and providing are implemented in at least one processing device comprising a processor coupled to a memory.

2. The method of claim 1 wherein the content comprises live video from at least one live video source.

3. The method of claim 1 wherein the content comprises at least a portion of a panoramic video comprising a combination of multiple live videos from respective ones of a plurality of video sources.

4. The method of claim 3 further comprising digitally cropping and scaling the panoramic video before providing a resulting version of the content for presentation to the viewer.

5. The method of claim 3 further comprising generating a plurality of different digitally cropped and scaled versions of the content from the panoramic video.

6. The method of claim 5 further comprising: providing a first one of the plurality of different digitally cropped and scaled versions of the content for presentation to the viewer; and responsive to input from the viewer, providing a second one of the plurality of different digitally cropped and scaled versions of the content for presentation to the viewer; wherein the second one of the plurality of different digitally cropped and scaled versions of the content is provided for presentation from a rewind point specified by the viewer.

7. The method of claim 1 wherein assigning an identifier to the object comprises assigning a player identifier to the object.

8. The method of claim 7 wherein providing additional information comprises providing the viewer with social media information associated with the player identifier.

9. The method of claim 7 wherein providing additional information comprises providing the viewer with a particular camera view associated with the object identifier.

10. The method of claim 9 wherein the particular camera view comprises a tracked player view in which a player corresponding to the player identifier is tracked over a plurality of frames and the camera view is cropped and scaled to increase an apparent size of the player in the tracked player view.

11. The method of claim 1 wherein the processing and assigning comprise: obtaining image coordinates for the object in each of a plurality of frames; determining a trajectory for the object over the frames based at least in part on the obtained image coordinates; classifying the object in each of at least a subset of the frames based at least in part on the trajectory; and assigning a player identifier to the classified object.

12. The method of claim 11 wherein obtaining image coordinates for the object in a given one of the frames comprises performing color blob detection based at least in part on a known team uniform color of one or more players.

13. The method of claim 11 wherein classifying the object comprises: defining an area of a given one of the frames using the obtained image coordinates for that frame; generating descriptors for the defined area using a dense feature set; applying the descriptors to a Gaussian mixture model to generate Fisher vectors; pooling the descriptors and associated Fisher vectors to form a spatial pyramid; and utilizing the spatial pyramid to classify the object within the given frame as corresponding to the player identifier.

14. The method of claim 1 wherein assigning an identifier to the object comprises: receiving input from the viewer associating the identified object with a particular player; and assigning a player identifier of the particular player to the identified object based at least in part on the received input.

15. The method of claim 1 wherein the processing and assigning comprise: detecting a preassigned player indicator on a uniform of a player corresponding to an identified object; assigning a player identifier to the identified object based at least in part on the detected preassigned player indicator; and processing the content to substantially remove the preassigned player indicator from the content before providing the content for presentation to the viewer.

16. The method of claim 1 further comprising: identifying players in the content; separating the content into a first portion comprising image portions associated with the players and a second portion comprising image portions associated with background; applying an anti-aliasing process to the second portion; and recombining the first and second portions; wherein the recombined first and second portions are provided for presentation to the viewer.

17. The method of claim 16 further comprising altering one or more advertisements of the first portion prior to recombining the first and second portions.

18. An article of manufacture comprising a computer-readable storage medium having computer program code embodied therein, wherein the computer program code when executed in at least one processing device causes said at least one processing device: to receive content from at least one source; to process the content to identify an object in the content; to assign an identifier to the object based on the processing; to permit viewer selection of the object identifier; and to provide additional information for presentation to a viewer of the content responsive to viewer selection of the object identifier.

19. An apparatus comprising: at least one processing device comprising a processor coupled to a memory; wherein said at least one processing device is configured to receive content from at least one source, to process the content to identify an object in the content, to assign an identifier to the object based on the processing, to permit viewer selection of the object identifier, and to provide additional information for presentation to a viewer of the content responsive to viewer selection of the object identifier.

20. The apparatus of claim 19 wherein said at least one processing device comprises at least one server.

21. A content delivery system comprising the apparatus of claim 19.