Back to results
US 11,232,585 B2Granted
Line-of-sight estimation device, line-of-sight estimation method, and program recording medium
AssigneeNEC Corporation
InventorYusuke Morishita
Life of this patent
Estimating the term…
Abstract
The purpose of the present invention is to improve the accuracy of line-of-sight estimation based on an image. A line-of-sight estimation device 100 comprises: an estimation unit 110 which estimates lines of sight from a face included in a face image, using a plurality of estimators; and a determination unit 120 which determines a line of sight for the face on the basis of first condition information including conditions relating to the capture of the face image, and a plurality of sets of second condition information, each including conditions associated with one of the plurality of estimators, and on the basis of a plurality of estimated lines of sight.
Claims15
A gaze estimation system, comprising:
- a memory that stores a set of instructions;
- and
- at least one processor configured to execute the set of instructions to:
- generate an estimation model associated with second condition information, the estimation model representing a result of performing a machine learning of estimating a gaze of a person by using face images which are captured in a plurality of imaging conditions that are different from each other, the second condition information including at least one of the plurality of imaging conditions;
- wherein
- the second condition information is comparable with first condition information that represents an imaging condition of a target face image in which a gaze of a person is estimated based on the estimation model;
- wherein the imaging condition includes a range of the gaze to be estimated, wherein
- the at least one processor is further configured to execute the set of instructions to:
- estimate the gaze in the target face image in the at least one of the imaging conditions indicated by the second condition information, based on the estimation model associated with the at least one of the imaging conditions indicated by the second condition information, and wherein
- the at least one processor is further configured to execute the set of instructions to:
- input a result of estimating the gaze in the target face image into a determination unit;
- wherein
- the determination unit determines the gaze in the target face image by weighting the result of estimating the gaze in the target face image, the result relating to the at least one of the imaging conditions indicated by the second condition information, based on a result of comparing the first condition information with the second condition information, wherein the weigh is determined based on an overlap ratio at which ranges of gazes overlap.
10 more elements
A gaze estimation method, comprising:
- by an information processing device:
- generating an estimation model associated with second condition information, the estimation model representing a result of performing a machine learning of estimating a gaze of a person by using face images which are captured in a plurality of imaging conditions that are different from each other, the second condition information including at least one of the plurality of imaging conditions;
- comparing the second condition information with first condition information that represents an imaging condition of a target face image in which a gaze of a person is estimated based on the estimation model, wherein the imaging condition includes a range of the gaze to be estimated;
- estimating the gaze in the target face image in the at least one of the imaging conditions indicated by the second condition information based on the estimation model associated with the at least one of the imaging conditions indicated by the second condition information;
- and
- inputting a result of estimating the gaze in the target face image into a determination unit;
- wherein
- the determination unit determines the gaze in the target face image by weighting the result of estimating the gaze in the target face image, the result relating to the at least one of the imaging conditions indicated by the second condition information, based on a result of comparing the first condition information with the second condition information, wherein the weigh is determined based on an overlap ratio at which ranges of gazes overlap.
5 more elements
A non-transitory computer-readable storage medium storing an estimation model, wherein
- the estimation model is generated by a computer associated with second condition information, the estimation model representing a result of performing a machine learning of estimating a gaze of a person by using face images which are captured in a plurality of imaging conditions that are different from each other, the second condition information including at least one of the plurality of imaging conditions,
- the second condition information is comparable with first condition information that represents an imaging condition of a target face image in which a gaze of a person is estimated based on the estimated model, wherein the imaging condition includes a range of the gaze to be estimated;
- the gaze is estimated in the target face image in the at least one of the imaging conditions indicated by the second condition information based on the estimation model associated with the at least one of the imaging conditions indicated by the second condition information, and
- a result of estimating the gaze in the target face image is input into a determination unit;
- wherein
- the determination unit determines the gaze in the target face image by weighting the result of estimating the gaze in the target face image, the result relating to the at least one of the imaging conditions indicated by the second condition information, based on a result of comparing the first condition information with the second condition information, wherein the weigh is determined based on an overlap ratio at which ranges of gazes overlap.
3 more elements
Dependencies are inferred from the claim text.