Synthesizing 3d hand pose based on multi-modal guided generative networks
AssigneeTencent America LLC
InventorsShih-Yao LIN, Kun WANG, Hui TANG, Chao HUANG, Lianyi HAN, Wei FAN
Life of this patent
Estimating the term…
Abstract
Systems and methods for obtaining hand images are provided. A method, performed by at least one processor that implements at least one network, includes obtaining a single source image, a three-dimensional (3D) hand pose of a first hand in the single source image, and a 3D target hand pose; and generating an image of a second hand, that has an appearance of the first hand and a pose of the 3D target hand pose, based on the single source image, the 3D hand pose, and the 3D target hand pose.
Claims20
A method performed by at least one processor that implements at least one network, the method comprising:
- obtaining a single source image, a three-dimensional (3D) hand pose of a first hand in the single source image, and a 3D target hand pose;
- and
- generating an image of a second hand, that has an appearance of the first hand and a pose of the 3D target hand pose, based on the single source image, the 3D hand pose, and the 3D target hand pose.
A system comprising:
- at least one processor, and
- memory comprising computer code, the computer code comprising network code configured to cause the at least one processor to implement at least one network that is configured to obtain a single source image, a three-dimensional (3D) hand pose of a first hand in the single source image, and a 3D target hand pose,
- wherein the network code comprises image generating code that is configured to cause the at least one processor to generate an image of a second hand, that has an appearance of the first hand and a pose of the 3D target hand pose, based on the single source image, the 3D hand pose, and the 3D target hand pose.
A non-transitory computer-readable medium storing computer code that is configured to, when executed by at least one processor, cause the at least one processor to:
- implement at least one network that is configured to obtain a single source image, a three-dimensional (3D) hand pose of a first hand in the single source image, and a 3D target hand pose;
- and
- generate an image of a second hand, that has an appearance of the first hand and a pose of the 3D target hand pose, based on the single source image, the 3D hand pose, and the 3D target hand pose.
Dependencies are inferred from the claim text.