Method and device for generating image effect of facial expression, and electronic device
View Patent ↗An expression image effect generation method and device, an electronic apparatus and a computer-readable storage medium are provided. The expression image effect generation method includes: acquiring a first image, where the first image includes a facial image; recognizing a current facial expression of the facial image; acquiring a previous facial expression; comparing the current facial expression with the previous facial expression to obtain a first comparison result; and generating, based on the first comparison result, an image effect corresponding to the first comparison result in the first image. In the method, an image effect can be generated by comparing the current facial expression and the previous facial expression, such that a user can quickly determine a degree of his expression.
1. An expression image effect generation method, comprising:
acquiring a first image being a video frame, wherein the first image comprises a facial image;
recognizing a current facial expression of the facial image in the video frame;
acquiring a previous facial expression, wherein the previous facial expression is a facial expression of the facial image in a previous video frame;
comparing the current facial expression with the previous facial expression to obtain a first comparison result, wherein the first comparison result indicates a relative degree of the current facial expression compared to the previous facial expression; and
generating an image effect in the video frame based on the relative degree of the current facial expression, wherein the image effect varies as the relative degree of the current facial expression changes.
2. The expression image effect generation method according to claim 1 , wherein the acquiring a first image, wherein the first image comprises a facial image, comprises:
acquiring a first video, wherein at least one video frame in the first video comprises the facial image.
3. The expression image effect generation method according to claim 1 , wherein the recognizing a current facial expression of the facial image comprises:
recognizing the facial image in the first image;
extracting a current facial expression feature from the facial image; and
recognizing the current facial expression based on the current facial expression feature.
4. The expression image effect generation method according to claim 1 , wherein the acquiring a previous facial expression comprises:
acquiring a default previous facial expression, wherein the default previous facial expression is preset before a facial expression of the facial image is recognized.
5. The expression image effect generation method according to claim 1 , wherein the acquiring a previous facial expression comprises:
acquiring the facial expression of the facial image at a previous sampling time.
6. The expression image effect generation method according to claim 1 , wherein the comparing the current facial expression with the previous facial expression to obtain a first comparison result comprises:
comparing a level of the current facial expression with a level of the previous facial expression, to obtain a relationship between the level of the current facial expression and the level of the previous facial expression.
7. The expression image effect generation method according to claim 1 , wherein the generating an image effect the video frame comprises:
acquiring, based on the first comparison result, an image effect profile corresponding to the first comparison result; and
rendering, based on the image effect profile, the image effect in the first image.
8. The expression image effect generation method according to claim 1 , wherein the acquiring a first image, wherein the first image comprises a facial image, comprises:
acquiring a first image, wherein the first image comprises at least two facial images.
9. The expression image effect generation method according to claim 8 , wherein the recognizing a current facial expression of the facial image comprises:
recognizing a current facial expression of each of the at least two facial images.
10. The expression image effect generation method according to claim 9 , wherein after the recognizing a current facial expression of each of the at least two facial images, the method further comprises:
acquiring, in a case that a facial expression of each of the at least two facial images is recognized for a first time, an image effect profile for the facial expression of each of the at least two facial images.
11. The expression image effect generation method according to claim 9 , wherein the acquiring a previous facial expression comprises:
acquiring a previous facial expression for the current facial expression of each of the at least two facial images.
12. The expression image effect generation method according to claim 11 , wherein the comparing the current facial expression with the previous facial expression to obtain a first comparison result comprises:
acquiring, for the facial expression of each of the at least two facial images, the current facial expression with the previous facial expression, to obtain a first comparison result for the facial expression of each of the at least two facial images.
13. The expression image effect generation method according to claim 11 , wherein the generating an image effect the video frame comprises:
generating, based on the first comparison result for the facial expression of each of the at least two facial images, an image effect corresponding to the first comparison result for the facial expression of each of the at least two facial images in the first image.
14. An electronic apparatus, comprising:
a memory storing non-transitory computer readable instructions; and
a processor, configured to run the computer-readable instructions, wherein the processor, when runs the computer-readable instructions, performs operations comprising:
acquiring a first image being a video frame, wherein the first image comprises a facial image;
recognizing a current facial expression of the facial image in the video frame;
acquiring a previous facial expression, wherein the previous facial expression is a facial expression of the facial image in a previous video frame;
comparing the current facial expression with the previous facial expression to obtain a first comparison result, wherein the first comparison result indicates a relative degree of the current facial expression compared to the previous facial expression; and
generating an image effect in the video frame based on the relative degree of the current facial expression, wherein the image effect varies as the relative degree of the current facial expression changes.
15. The apparatus according to claim 14 , wherein the acquiring a first image, wherein the first image comprises a facial image, comprises:
acquiring a first video, wherein at least one video frame in the first video comprises the facial image.
16. The apparatus according to claim 14 , wherein the recognizing a current facial expression of the facial image comprises:
recognizing the facial image in the first image;
extracting a current facial expression feature from the facial image; and
recognizing the current facial expression based on the current facial expression feature.
17. The apparatus according to claim 14 , wherein the acquiring a previous facial expression comprises:
acquiring a default previous facial expression, wherein the default previous facial expression is preset before a facial expression of the facial image is recognized.
18. The apparatus according to claim 14 , wherein the acquiring a previous facial expression comprises:
acquiring the facial expression of the facial image at a previous sampling time.
19. The apparatus according to claim 14 , wherein the comparing the current facial expression with the previous facial expression to obtain a first comparison result comprises:
comparing a level of the current facial expression with a level of the previous facial expression, to obtain a relationship between the level of the current facial expression and the level of the previous facial expression.
20. A non-transitory computer-readable storage medium comprising non-transitory computer-readable instructions stored thereon, wherein the non-transitory computer-readable instructions, when executed by a computer, cause the computer to perform operations comprising:
acquiring a first image being a video frame, wherein the first image comprises a facial image;
recognizing a current facial expression of the facial image in the video frame;
acquiring a previous facial expression, wherein the previous facial expression is a facial expression of the facial image in a previous video frame;
comparing the current facial expression with the previous facial expression to obtain a first comparison result, wherein the first comparison result indicates a relative degree of the current facial expression compared to the previous facial expression; and
generating an image effect in the video frame based on the relative degree of the current facial expression, wherein the image effect varies as the relative degree of the current facial expression changes.