摘要:
A line-of-sight detection device includes a camera, a light source that causes corneal reflection, and an image processing device that obtains information on a pupil shape of an eye, a position of a pupil center, and a position of the cornea reflection and obtains a line of sight of a subject. The image processing device includes a first optical axis acquisition unit that acquires a first optical axis candidate using the position of the pupil center and the position of the cornea reflection, a second optical axis acquisition unit that obtains a second optical axis candidate using the pupil shape, an optical axis information acquisition unit that generates information for acquiring the optical axis, a calibration information unit that provides calibration information, and a visual axis acquisition unit that obtains a visual axis as a line of sight by calibrating an optical axis using the calibration information.
摘要:
A pupil detection light source device includes an aperture disposed to face a subject to allow light from a pupil of the subject to pass therethrough, an illumination light for obtaining a light pupil image toward the subject from the inside or the vicinity of the aperture when seen from the subject, and an illumination light for forming a reflection image with glasses obtained by canceling a reflection image with glasses of the light pupil image using an image difference toward the subject from the inside or the vicinity of the aperture when seen from the subject.
摘要:
A pupil detection method includes a step of acquiring images by taking a facial image of a subject using each of a left camera and a right camera, a step of extracting one or more image candidate points serving as candidates for pupils from the images, a step of extracting points corresponding to a combination of image candidate points corresponding to the same point in a three-dimensional space as space candidate points, a step of selecting a pair of two space candidate points from the extracted space candidate points and calculating a distance between the selected pair, a step of excluding a pair of space candidate points where the calculated distance is not within a specified range, and a step of determining one or more pairs of space candidate points from the pairs not excluded and determining that a pair of pupils of the subject exist at the positions.
摘要:
A pupil detection method includes a step of acquiring images by taking a facial image of a subject using each of a left camera and a right camera, a step of extracting one or more image candidate points serving as candidates for pupils from the images, a step of extracting points corresponding to a combination of image candidate points corresponding to the same point in a three-dimensional space as space candidate points, a step of selecting a pair of two space candidate points from the extracted space candidate points and calculating a distance between the selected pair, a step of excluding a pair of space candidate points where the calculated distance is not within a specified range, and a step of determining one or more pairs of space candidate points from the pairs not excluded and determining that a pair of pupils of the subject exist at the positions.
摘要:
A corneal reflection position estimation system according to an embodiment includes: an image acquisition unit configured to continuously acquire an image of an eye of a subject by controlling a camera equipped with a light source, the image acquisition unit being configured to acquire a plurality of pupil images taken by using light from the light source and then acquire one unilluminated image taken without using light from the light source; and an estimation unit configured to calculate positions of corneal reflection or corneal sphere centers, which correspond to the plurality of pupil images, as reference positions, calculate a movement vector of the corneal reflection or the corneal sphere center based on a plurality of the reference positions, and estimate the corneal reflection position in the unilluminated image based on at least the movement vector.