Creating a mixed-reality video based upon tracked skeletal features
AssigneeMicrosoft Technology Licensing, LLC
InventorsJohn MCELMURRAY, Gustaf PETTERSSON, Karim BENCHEMSI
Life of this patent
Estimating the term…
Abstract
Technologies related to generating mixed reality video are described herein. Video frames capture a human, and positions of skeletal features of the human are estimated based solely upon the video frames. A mixed reality video is generated, wherein the mixed reality video includes the video combined with an animation, wherein the animation is based upon movement of the human in the video as represented by the estimated positions of the skeletal features of the human.
Claims20
A computing system comprising:
- a processor;
- and
- memory storing instructions that, when executed by the processor, cause the processor to perform acts comprising:
- estimating positions of skeletal features of a human body captured in frames of video, wherein the positions of the skeletal features of the human body in the frames are estimated based solely upon the frames;
- generating mixed reality video frames based upon the estimated positions of the skeletal features of the human body, wherein the mixed reality video frames include the frames combined with an animation;
- and
- storing the mixed reality video frames in a computer-readable storage device.
A method executed on a computing device, the method comprising:
- estimating positions of skeletal features of a human in frames of video, wherein the positions of the skeletal features are estimated based solely upon the frames of the video;
- and
- generating a mixed reality video based upon the frames of the video and the estimated positions of the skeletal features of the human, wherein the mixed reality video includes an animation combined with the frames of the video, wherein the animation corresponds to movement of the human as represented by the estimated positions of the skeletal features of the human in the frames of the video.
A computer-readable storage medium comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:
- receiving a video file, wherein the video file comprises frames, and further wherein the frames include a human body;
- estimating positions of skeletal features in the frames based solely upon the frames;
- receiving a selection of an animation;
- and
- based upon the selection of the animation, the frames, and the estimated positions of the skeletal features in the frames, generating a mixed reality video file, wherein the mixed reality video file includes the frames combined with the animation, wherein the animation reflects movement of the human body as represented by the estimated positions of the skeletal features in the frames.
Dependencies are inferred from the claim text.