US 12,216,292 B2
Automated video capture and composition system
Matthew Hanover, Los Angeles, CA (US); and Richard Zhuang, Los Angeles, CA (US)
Assigned to Snap Inc., Santa Monica, CA (US)
Filed by Snap Inc., Santa Monica, CA (US)
Filed on Apr. 28, 2022, as Appl. No. 17/731,823.
Application 17/731,823 is a continuation of application No. 16/585,603, filed on Sep. 27, 2019, granted, now 11,320,667, issued on May 3, 2022.
Prior Publication US 2022/0252894 A1, Aug. 11, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G02B 27/64 (2006.01); G02B 27/01 (2006.01); H04N 1/00 (2006.01)
CPC G02B 27/64 (2013.01) [G02B 27/017 (2013.01); H04N 1/00246 (2013.01); G02B 2027/0138 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method of producing a video composition, comprising:
capturing frames of raw video with a camera coupled to a wearable device in motion along a series of positions relative to a physical environment;
collecting motion data with an inertial measurement unit coupled to the wearable device;
computing a stabilized output path based on the collected motion data, such that the stabilized output path approximates the series of positions, wherein computing the stabilized output path comprises:
(a) estimating a camera orientation associated with each of one or more of the frames of raw video, and (b) establishing one or more waypoints along the stabilized output path based on the camera orientations; and
producing a video composition based on the stabilized output path, wherein producing the video composition comprises adjusting the one or more frames of raw video relative to the waypoints, such that at least a portion of the video composition appears stable relative to a reference plane.