Facial performance capture in an uncontrolled environment

Patent metadata

Field	Value
Publication number	US-11049332-B2
Application number	US-202016808110-A
Country	US
Kind code	B2
Filing date	Mar 3, 2020
Priority date	Mar 7, 2019
Publication date	Jun 29, 2021
Grant date	Jun 29, 2021

Abstract

A method of transferring a facial expression from a subject to a computer generated character that includes: receiving a plate with an image of the subject's facial expression and an estimate of intrinsic parameters of a camera used to film the plate; generating a three-dimensional parameterized deformable model of the subject's face where different facial expressions of the subject can be obtained by varying values of the model parameters; solving for the facial expression in the plate by executing a deformation solver to solve for at least some parameters of the deformable model with a differentiable renderer and shape-from-shading techniques, using as inputs, the three-dimensional parameterized deformable model, estimated intrinsic camera parameters, estimated lighting conditions and albedo estimates over a series of iterations to infer geometry of the facial expression and generate an intermediate facial; generating, from the intermediate facial mesh, refined albedo estimates for the deformable model; and solving for the facial expression in the plate by executing the deformation solver using the intermediate facial mesh, the estimated intrinsic camera parameters, the estimated lighting conditions and the refined albedo estimates as inputs over a series of iterations to infer geometry of the facial expression and generate a final facial mesh using the set of parameter values of the deformable model which result in a facial expression that more closely matches the expression of the subject in the plate than does the intermediate facial mesh.

First claim

What is claimed is:

1. A method of transferring a facial expression from a subject to a computer generated character where the subject was filmed in an uncontrolled environment with low-frequency, static illumination, the method comprising:
receiving a plate with an image of the facial expression of the subject and an estimate of intrinsic parameters of a camera used to film the plate;
generating a three-dimensional parameterized deformable model of a face of the subject, the three-dimensional parameterized deformable model comprising a plurality of model parameters where different facial expressions of the subject can be obtained by varying values of the plurality of model parameters;
generating, from the plate, an estimate of lighting conditions that illuminate the subject in the plate and albedo estimates for the three-dimensional parameterized deformable model;
initially solving for the facial expression in the plate by executing a deformation solver to solve for at least some parameters of the three-dimensional parameterized deformable model with a differentiable renderer and shape-from-shading techniques, using as inputs, the three-dimensional parameterized deformable model, the estimated intrinsic camera parameters, the estimated lighting conditions and the albedo estimates over a first series of iterations to infer geometry of the facial expression of the subject in the plate and generate an intermediate facial mesh using a set of parameter values of the three-dimensional parameterized deformable model which result in an intermediate facial expression that approximates the facial expression of the subject in the plate;
generating, from the intermediate facial mesh, refined albedo estimates for the three-dimensional parameterized deformable model; and
subsequently solving for the facial expression in the plate by executing the deformation solver using the intermediate facial mesh, the estimated intrinsic camera parameters, the estimated lighting conditions and the refined albedo estimates as inputs over a second series of iterations to infer geometry of the facial expression of the subject in the plate and generate a final facial mesh using the set of parameter values of the three-dimensional parameterized deformable model which result in a final facial expression that more closely matches the facial expression of the subject in the plate than the intermediate facial mesh.

2. The method of claim 1 , wherein, the three-dimensional parameterized deformable model comprises a plurality of blendshapes, each blendshape in the plurality of blendshapes representing a different facial expression of the subject and wherein, the set of parameter values of the three-dimensional parameterized deformable model comprise a set of blendshape weight values, one blendshape weight value per each blendshape.

3. The method of claim 2 , wherein, the set of parameter values of the three-dimensional parameterized deformable model further comprise rotation and translation values that represent a rigid adjustment of a head of the subject.

4. The method of claim 3 , wherein the step of solving for the facial expression in the plate to generate the intermediate facial mesh includes, in order: solving for shape weight values from rotoscoping splines and 2D landmarks on the plate, solving for the shape weights from the plate, and then solving the shape weight, the rotation and the translation values using the shape-from-shading techniques.

5. The method of claim 3 , wherein, the computer-generated character includes a head that is sized or shaped differently than the head of the subject and the set of parameter values of the three-dimensional parameterized deformable model further comprise a value for a delta vector that represents a per vertex displacement used in transferring the facial expression of the subject to the computer-generated character.

6. The method of claim 3 , wherein, the plate is an image comprising more than a million pixels with each pixel having a particular RGB value and wherein, during each iteration, the differentiable renderer generates a rendering of the three-dimensional parameterized deformable model and the deformation solver tries to minimize differences between the RGB values of the plate and RGB values of corresponding pixels in the rendered version of the three-dimensional parameterized deformable model.

7. The method of claim 6 , wherein, an initial iteration of the initially solving for the facial expression in the plate comprises:
rendering an initial facial mesh, generated from the three-dimensional parameterized deformable model, representing a neutral expression of the subject;
trying to minimize differences between the RGB values of the plate and RGB values of the rendered initial facial mesh representing the neutral expression; and
generating an updated facial mesh in which the facial expression of the subject in the updated facial mesh more closely matches the facial expression of the subject in the plate than the facial expression of the initial facial mesh.

8. The method of claim 7 , wherein, for each iteration, an output of current iteration is closer to actual representation of the facial expression of the subject in the plate than an output of the previous iteration.

9. The method of claim 1 , wherein the albedo estimates are generated and then refined at different vertices of the three-dimensional parameterized deformable model.

10. The method of claim 2 , further comprising generating a visual effect from the set of blendshape weight values and incorporating the visual effect into one or more of an animation, a movie, a video, a video game, or virtual or augmented reality content.

11. The method of claim 1 , wherein, the three dimensional parameterized deformable model of the face of the subject includes a facial rig generated by an artist who modifies a neutral expression of a generic human facial rig to better match the face of the subject.

12. A method of transferring a facial expression from a subject in a previously filmed performance to a computer generated character where the subject was filmed in an uncontrolled environment with low-frequency, static illumination, the method comprising:
receiving digital footage of the previously filmed performance including a plurality of sequentially ordered plates each of which includes an image of the facial expression of the subject and receiving an estimate of intrinsic parameters of a camera used to film the plurality of plates;
generating a three-dimensional parameterized deformable model of a face of the subject, the three-dimensional parameterized model comprising a plurality of model parameters where different facial expressions of the subject can be obtained by varying values of the plurality of model parameters;
generating, from one or more plates in the plurality of plates, an estimate of lighting conditions that illuminate the subject in each plate and albedo estimates at different vertices of the three-dimensional parameterized deformable model;
generating a computer model of the previously filmed performance by, for each individual plate in the plurality of sequentially ordered plates, processing the individual plate independently of other plates in the plurality of plates to solve for the facial expression in the plate being processed by:
executing a deformation solver to initially solve for at least some parameters of the three-dimensional parameterized deformable model with a differentiable renderer and shape-from-shading techniques, using as inputs, the three-dimensional parameterized deformable model, the estimated intrinsic camera parameters, the estimated lighting conditions and the albedo

Assignees

• Lucasfilm Entertainment Co Ltd

Inventors

• Loper Matthew
• Grabli Stéphane
• Bhat Kiran

Classifications

• G06T2207/30201

  Face · CPC title

• G06T13/40

  of characters, e.g. humans, animals or virtual beings · CPC title

• G06T7/80

  Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration · CPC title

• G06T19/20 Primary

  Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts · CPC title

• G06T2219/2021

  Shape modification · CPC title