Creating a non-riggable model of a face of a person
A method for creating a non-riggable model of a face of a person, the method includes obtaining video and depth information regarding the face of the person, wherein different images of the video are acquired by a camera at different camera locations; and for each image of the different images repeating the steps of: separating face information from background information; determining translation and rotation parameters that represent the different camera locations; and generating the non-riggable model of the face of the person based on the face information and the translation and rotation parameters.
1. A method for creating a non-riggable model of a face of a person, the method comprises:
obtaining video and depth information regarding the face of the person, wherein different images of the video are acquired by a camera at different camera locations;
for each image of the different images repeating the steps of:
separating face information from background information;
determining translation and rotation parameters that represent the different camera locations;
generating the non-riggable model of the face of the person based on the face information and the translation and rotation parameters;
wherein the different camera locations comprise multiple sets of camera locations, each set comprises camera locations that are proximate to each other, wherein the determining of the translation and rotation parameters comprises calculating local translation and rotation parameters for each one of the sets of camera locations; and
determining shared-base translation and rotation parameters between a certain camera location and each one of the local translations and rotation parameters.
2. The method according to claim 1 , wherein the determining of the translation and rotation parameters is based on only a part of the different images.
3. The method according to claim 1 , wherein the determining of the translation and rotation parameters is based on only a part of face landmarks of the different images.
4. The method according to claim 1 , wherein the determining of the translation and rotation parameters is based on face portions of different 3D points clouds obtained for the different images.
5. The method according to claim 1 , wherein the determining of the translation and rotation parameters comprises applying a random sample consensus process.
6. The method according to claim 1 , wherein for each image of the different images also repeating the steps of: semantically segmenting the image to provide image segments, wherein at least some of the image segments correspond to face landmarks; and generating a three dimensional (3D) points cloud that correspond to the face landmarks of the image.
7. A non-transitory computer readable medium for creating a non-riggable model of a face of a person, the non-transitory computer readable medium stores instructions that once executed by a processor cause the processor to execute steps, the steps comprising:
obtaining video and depth information regarding the face of the person, wherein different images of the video are acquired by a camera at different camera locations;
for each image of the different images repeating the steps of:
separating face information from background information; determining translation and rotation parameters that represent the different camera locations;
generating the non-riggable model of the face of the person based on the face information and the translation and rotation parameters; and
wherein the different camera locations comprise multiple sets of camera locations, each set comprises camera locations that are proximate to each other, wherein the determining of the translation and rotation parameters comprises calculating local translation and rotation parameters for each one of the sets of camera locations; and
determining shared-base translation and rotation parameters between a certain camera location and each one of the local translations and rotation parameters.
8. The non-transitory computer readable medium according to claim 7 , wherein the determining of the translation and rotation parameters is based on only a part of the different images.
9. The non-transitory computer readable medium according to claim 7 , wherein the determining of the translation and rotation parameters is based on only a part of face landmarks of the different images.
10. The non-transitory computer readable medium according to claim 7 , wherein the determining of the translation and rotation parameters is based on face portions of different 3D points clouds obtained for the different images.
11. The non-transitory computer readable medium according to claim 7 , wherein the determining of the translation and rotation parameters comprises applying a random sample consensus process.
12. The non-transitory computer readable medium according to claim 7 , that stores instructions for repeating, for each image of the different images: semantically segmenting the image to provide image segments, wherein at least some of the image segments correspond to face landmarks; and generating a three dimensional (3D) points cloud that correspond to the face landmarks of the image.