Granted Patent B2

US 12,632,107 · App. 19/016,036 · Granted May 19, 2026

Inventors: Hiromichi Godo (Isehara, JP); Yoshiyuki Kurokawa (Sagamihara, JP); Seiko Inoue (Atsugi, JP); Kazuaki Ohshima (Atsugi, JP); Shunpei Yamazaki (Setagaya, JP)

Assignee: Semiconductor Energy Laboratory Co., Ltd.

G06F3/013G06T1/20G06T7/75G06V10/141G06V40/193H04N23/611H04N23/617H04N23/90G06T2207/10048G06T2207/20081G06T2207/30041G06T2207/30201

Abstract

Claims (36)

1 . An electronic device comprising:

a display portion comprising a first camera;

a second camera; and

an image processing portion,

wherein a pixel of the display portion comprises a transistor and an organic EL device,

wherein the transistor is configured to drive the pixel,

wherein the transistor comprises an oxide semiconductor layer,

wherein the second camera is positioned in a region not overlapping with the display portion,

wherein the first camera is configured to generate a first image of a subject,

wherein the second camera is configured to generate a second image of the subject,

wherein the image processing portion comprises a generator including a neural network, the neural network being configured to perform learning using training data,

wherein the training data comprises an image comprising a face of a person,

wherein the image processing portion is configured to deblur the first image when the first image is input to the generator and track a gaze of the subject on the basis of the second image,

wherein the display portion comprises a second region overlapping with the first camera, and a third region not overlapping with the first camera, and

wherein an arrangement of pixels of the second region is different from an arrangement of pixels of the third region.

2 . The electronic device according to claim 1 , further comprising a light source emitting infrared light,

wherein the light source is positioned in the region not overlapping with the display portion,

wherein the light source is used for detection of the gaze of the subject, and

wherein the gaze of the subject is tracked by repeating the detection of the gaze of the subject.

3 . The electronic device according to claim 1 , wherein the first camera is positioned behind the pixel when seen from the subject.

4 . The electronic device according to claim 1 , wherein the first camera is positioned in a first region comprising the pixel when seen from the subject.

5 . An electronic device comprising:

a display portion comprising a first camera;

a second camera; and

an image processing portion,

wherein a pixel of the display portion comprises a transistor and an organic EL device,

wherein the transistor is configured to drive the pixel,

wherein the transistor comprises an oxide semiconductor layer,

wherein the second camera is positioned in a region not overlapping with the display portion,

wherein the first camera is configured to generate a first image of a subject,

wherein the second camera is configured to generate a second image of the subject,

wherein the image processing portion comprises a generator including a neural network, the neural network being configured to perform learning using training data,

wherein the training data comprises an image comprising a face of a person,

wherein the image processing portion is configured to denoise the first image when the first image is input to the generator and track a gaze of the subject on the basis of the second image,

wherein the display portion comprises a second region overlapping with the first camera, and a third region not overlapping with the first camera, and

wherein an arrangement of pixels of the second region is different from an arrangement of pixels of the third region.

Priority Claims (1)

Continuity (2)

References Cited (88)