Granted Patent B2

US 12,327,299 · App. 17/426,840 · Granted Jun 10, 2025

Inventors: Shaohui Lyu (Beijing, CN); Hui Yang (Beijing, CN); Guangyao Ni (Beijing, CN)

Assignee: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD.

G06T11/00G06T3/40G06V10/70G06V40/161G06V40/168G06V40/174

Abstract

Claims (77)

1. A computer-implemented method for generating image effects, comprising:

acquiring a first image, the first image comprising a facial image;

recognizing a facial expression of the facial image;

performing, in response to recognizing the facial expression as a first facial expression, a first processing on the facial image to acquire a first facial image, wherein the performing the first processing on the facial image to acquire a first facial image comprises:

computing a position of a pixel in the facial image based on segmenting the facial image from the first image and enlarging the segmented facial image,

assigning a color value to the pixel in the segmented and enlarged facial image based at least in part on a color value of the pixel in the facial image, and

generating a new facial image as the first facial image based on the segmented and enlarged facial image with assigned color values; and

superposing the first facial image on the facial image to generate a first image effect.

2. The method for processing a facial expression image according to claim 1 , wherein the acquiring a first image, the first image comprising a facial image comprises:

acquiring a first video, wherein at least one of video frames in the first video comprises the facial image.

3. The method for processing a facial expression image according to claim 1 , wherein the recognizing a facial expression of the facial image comprises:

recognizing the facial image in the first image;

extracting a feature of the facial expression from the facial image; and

recognizing the facial expression based on the feature of the facial expression.

4. The method for processing a facial expression image according to claim 1 , wherein the performing, in response to recognizing the facial expression as a first facial expression, a first processing on the facial image to acquire a first facial image comprises:

acquiring a processing configuration file corresponding to the first facial expression, in response to recognizing the racial expression as the first facial expression; and

performing the first processing on the facial image based on the processing configuration file to acquire the first facial image.

5. The method for processing a facial expression image according to claim 4 , wherein the acquiring a processing configuration file corresponding to the first facial expression, in response to recognizing the facial expression as the first facial expression comprises:

recognizing the facial expression as the first facial expression; and

acquiring the processing configuration file corresponding to the first facial expression, in a case that a level of the first facial expression reaches a preset level.

6. The method for processing a facial expression image according to claim 4 , wherein the acquiring a processing configuration file corresponding to the first facial expression, in response to recognizing the facial expression as the first facial expression comprises:

recognizing the facial expression as the first facial expression;

acquiring the processing configuration file corresponding to the first facial expression;

determining a level of the first facial expression; and

setting a processing parameter in the processing configuration file based on the level of the first facial expression.

7. The method for processing a facial expression image according to claim 4 , wherein the performing the first processing on the facial image based on the processing configuration file to acquire the first facial image comprises:

segmenting the facial image from the first image; and

enlarging the facial image segmented from the first image based on the processing configuration file, to acquire an enlarged facial image.

8. The method for processing a facial expression image according to claim 1 , wherein the covering the first facial image at a position of the facial image, to acquire a first image effect comprises:

acquiring a first positioning feature point on the first facial image and a second positioning feature point on the facial image; and

covering the first facial image on the facial image, and causing the first positioning feature point and the second positioning feature point to coincide with each other, to acquire the first image effect.

9. The method for processing a facial expression image according to claim 1 , wherein the acquiring a first image, the first image comprising a facial image comprises:

acquiring the first image, the first image comprising at least two facial images.

10. The method for processing a facial expression image according to claim 9 , wherein the recognizing a facial expression of the facial image comprises:

respectively recognizing facial expressions of the at least two facial images.

11. The method for processing a facial expression image according to claim 10 , wherein the performing, in response to recognizing the facial expression as a first facial expression, a first processing on the facial image to acquire a first facial image comprises:

performing, in response to recognizing at least one of the facial expressions as the first facial expression, the first processing on the facial image corresponding to the first facial expression to acquire at least one first facial image.

12. The method for processing a facial expression image according to claim 11 , wherein the performing, in response to recognizing at least one of the facial expressions as the first facial expression, the first processing on the facial image corresponding to the first facial expression to acquire at least one first racial image comprises:

in response to recognizing the at least one or the facial expressions as the first facial expression, acquiring a first processing configuration file corresponding to the first facial expression of the facial image; and

performing the first processing on the facial image corresponding to the first facial expression, based on the first processing configuration file to acquire the at least one first facial image.

13. The method for processing a facial expression image according to claim 11 , wherein the covering the first facial image at a position of the facial image, to acquire a first image effect comprises:

covering the at least one first facial image at a position of the facial image corresponding to the at least one first facial image, to acquire the first image effect.

14. An electronic device, comprising:

at least one processor; and

at least one memory configured for storing at least one program, wherein the at least one program, when executed by the at least one processor, causes the at least one processor to perform operations comprising:

acquiring a first image, the first image comprising a facial image;

recognizing a facial expression of the facial image;

performing, in response to recognizing the facial expression as a first facial expression, a first processing on the facial image to acquire a first facial image, wherein the performing the first processing on the facial image to acquire a first facial image comprises:

computing a position of a pixel in the facial image based on segmenting the facial image from the first image and enlarging the segmented facial image,

assigning a color value to the pixel in the segmented and enlarged facial image based at least in part on a color value of the pixel in the facial image, and

generating a new facial image as the first facial image based on the segmented and enlarged facial image with assigned color values; and

superposing the first facial image on the facial image to generate a first image effect.

15. A non-transitory computer readable storage medium configured to store non-transitory computer readable instructions, the non-transitory computer readable instructions, when executed by a computer, causing the computer to perform operations comprising:

acquiring a first image, the first image comprising a facial image;

recognizing a facial expression of the facial image;

performing, in response to recognizing the facial expression as a first facial expression, a first processing on the facial image to acquire a first facial image, wherein the performing the first processing on the facial image to acquire a first facial image comprises:

computing a position of a pixel in the facial image based on segmenting the facial image from the first image and enlarging the segmented facial image,

assigning a color value to the pixel in the segmented and enlarged facial image based at least in part on a color value of the pixel in the facial image, and

generating a new facial image as the first facial image based on the segmented and enlarged facial image with assigned color values; and

superposing the first facial image on the facial image to generate a first image effect.

16. The method for processing a facial expression image according to claim 12 , wherein the covering the first facial image at a position of the facial image, to acquire a first image effect comprises:

covering the at least one first facial image at a position of the facial image corresponding to the at least one first facial image, to acquire the first image effect.

17. The electronic device according to claim 14 , wherein the recognizing a facial expression of the facial image comprises:

recognizing the facial image in the first image;

extracting a feature of the facial expression from the facial image; and

recognizing the facial expression based on the feature of the facial expression.

18. The electronic device according to claim 14 , wherein the performing, in response to recognizing the facial expression as a first facial expression, a first processing on the facial image to acquire a first facial image comprises:

acquiring a processing configuration file corresponding to the first facial expression, in response to recognizing the facial expression as the first facial expression; and

performing the first processing on the facial image based on the processing configuration file to acquire the first facial image.

19. The electronic device according to claim 18 , wherein the acquiring a processing configuration file corresponding to the first facial expression, in response to recognizing the facial expression as the first facial expression comprises:

recognizing the facial expression as the first facial expression; and

acquiring the processing configuration file corresponding to the first facial expression, in a case that a level of the first facial expression reaches a preset level.

20. The electronic device according to claim 18 , wherein the acquiring a processing configuration file corresponding to the first facial expression, in response to recognizing the facial expression as the first facial expression comprises:

recognizing the facial expression as the first facial expression;

acquiring the processing configuration file corresponding to the first facial expression;

determining a level of the first facial expression; and

setting a processing parameter in the processing configuration file based on the level of the first facial expression.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (41)