Systems and methods for automatic multi-dance transition
AssigneeKingstar Technologies, Inc.
InventorHee-Chul Han
Life of this patent
Estimating the term…
Abstract
Video processing systems and methods are disclosed. An exemplary video processing method includes receiving a first performance video and a second performance video, and estimating, by at least one processor, pose information in each of the first performance video and the second performance video. The video processing method further includes determining, by the at least one processor, a point of transition from the first performance video to the second performance video based on a pose distance calculated using the estimated pose information; and generating, by the at least one processor, a third performance video by transitioning the first performance video to the second performance video at the point of transition.
Claims18
A video processing method, comprising:
- receiving a first performance video and a second performance video;
- estimating, by at least one processor, pose information in each of the first performance video and the second performance video;
- determining a music pattern in each of the first performance video and the second performance video based on audio data in the corresponding performance video;
- determining, by the at least one processor, a point of transition from the first performance video to the second performance video based on a pose distance calculated using the estimated pose information and by matching music patterns of the first performance video and the second performance video;
- and
- generating, by the at least one processor, a third performance video by transitioning the first performance video to the second performance video at the point of transition.
A video processing system, comprising:
- a storage device configured to receive a first performance video and a second performance video;
- and
- at least one processor configured to:
- estimate pose information in each of the first performance video and the second performance video;
- determining a music pattern in each of the first performance video and the second performance video based on audio data in the corresponding performance video;
- determine a point of transition from the first performance video to the second performance video based on a pose distance calculated using the estimated pose information and by matching music patterns of the first performance video and the second performance video;
- and
- generate a third performance video by transitioning the first performance video to the second performance video at the point of transition.
A non-transitory computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by at least one processor, causes the at least one processor to perform a video processing method, comprising:
- receiving a first performance video and a second performance video;
- estimating pose information in each of the first performance video and the second performance video;
- determining a music pattern in each of the first performance video and the second performance video based on audio data in the corresponding performance video;
- determining a point of transition from the first performance video to the second performance video based on a pose distance calculated using the estimated pose information and by matching music patterns of the first performance video and the second performance video;
- and
- generating a third performance video by transitioning the first performance video to the second performance video at the point of transition.
Dependencies are inferred from the claim text.