Method for generating image having metal texture, and model training method

1 . A method of generating an image with metallic texture, comprising:

acquiring a first video; and

inputting the first video into a pre-trained metallic texture image generation model to obtain a second video, wherein each frame of images in the second video is an image with metallic texture;

wherein the metallic texture image generation model is obtained by updating model parameters of an initial model based on a plurality of sample image pairs,

the plurality of sample image pairs are determined based on a plurality of second sample images and a plurality of target sample images, and

the plurality of second sample images are determined based on a plurality of first sample images with metallic texture, wherein the plurality of first sample images are determined based on the plurality of target sample images, and wherein a difference between a position of a human face in each of the plurality of second sample images and a position of a human face in a corresponding target sample image among the plurality of target sample images is less than a preset value.

2 . The method according to claim 1 , wherein each second sample image is obtained by inputting a grayscale image corresponding to a target sample image into a second network model;

the second network model is obtained by training a first network model based on a third sample image and the grayscale image corresponding to the target sample image; and

the third sample image is determined based on a first sample image, and

wherein the third sample image is based on an alignment of second facial keypoints in the first sample image and first facial keypoints in the target sample image.

3 . The method according to claim 2 , wherein the sample image pair is derived based on a fourth sample image and the target sample image; and

the fourth sample image is obtained by fusing the second sample image and the third sample image, and

wherein the fourth sample image is obtained by fusing the second sample image and the corresponding third sample image based on a position of a human face and a first mask image; and the position of the human face and the first mask image are obtained by performing a human face segmentation on the target sample image.

4 . The method according to claim 3 , wherein the fourth sample image is obtained by fusing the second sample image and the corresponding third sample image based on the position of the human face and a second mask image; and

the second mask image is obtained by performing a morphological manipulation process and/or a Gaussian blur process on the first mask image.

5 . The method according to claim 1 , wherein each sample image pair is derived based on a second sample image and a processed image; and

the processed image is obtained by performing an attribute information adjustment process on a target sample image.

6 . The method according to claim 2 , wherein the second network model comprises a plurality of third network models resulting from training the first network model;

the second sample image is obtained by performing a transparency blending process on a plurality of fifth sample images; and

the plurality of fifth sample images are obtained by processing the grayscale image by the plurality of third network models, respectively.

7 . A non-transitory computer-readable storage medium with computer-executable instructions stored thereon, wherein the computer-executable instructions, when executed by a processor, are configured to implement the method of claim 1 .

8 . A method of training a metallic texture image generation model, comprising:

determining a first sample image corresponding to each target sample image, wherein the first sample image is an image with metallic texture;

determining a second sample image corresponding to each target sample image according to the first sample image corresponding to each target sample image, wherein a difference between a position of a human face in the second sample image and a position of a human face in the corresponding target sample image is less than a preset value; and

determining a plurality of sample image pairs based on each second sample image and each target sample image, and updating model parameters of an initial metallic texture image generation model based on the plurality of sample image pairs to obtain the metallic texture image generation model.

9 . The method according to claim 8 , wherein determining the second sample image corresponding to the target sample image according to the first sample image comprises:

determining, according to each first sample image, a third sample image corresponding to each first sample image, wherein a difference between a position of a human face in the third sample image and the position of the human face in the target sample image is less than a difference between a position of a human face in the first sample image and the position of the human face in the target sample image;

training a first network model according to each third sample image and a grayscale image corresponding to each target sample image to obtain a second network model; and

processing the grayscale image corresponding to the target sample image by the second network model to obtain the second sample image.

10 . The method according to claim 9 , wherein determining, according to the first sample image, the third sample image corresponding to the first sample image comprises:

extracting a plurality of first facial keypoints in the target sample image and a plurality of second facial keypoints in the first sample image, respectively; and

aligning second facial keypoints in the first sample image with first facial keypoints in the target sample image to obtain the third sample image.

11 . The method according to claim 9 , wherein determining the sample images pairs based on the second sample image and the target sample image comprises:

fusing the second sample image and the third sample image to obtain a fourth sample image; and

determining the sample image pair according to the fourth sample image and the target sample image.

12 . The method according to claim 11 , wherein the fusing the second sample image and the third sample image to obtain the fourth sample image comprises:

performing a human face segmentation on the target sample image to obtain a position of a human face and a first mask image; and

fusing the second sample image and the corresponding third sample image according to the position of the human face and the first mask image to obtain the fourth sample image, and

wherein the fusing the second sample image and the corresponding third sample image according to the position of the human face and the first mask image to obtain the fourth sample image comprises:

performing a morphological manipulation process and/or a Gaussian blur process on the first mask image to obtain a second mask image; and

fusing the second sample image and the corresponding third sample image according to the position of the human face and the second mask image to obtain the fourth sample image.

13 . The method according to claim 8 , wherein determining the sample image pair based on the second sample image and the target sample image comprises:

performing an attribute information adjustment process on the target sample image to obtain a processed image; and

determining the second sample image and the processed image as a sample image pair.

14 . The method according to claim 9 , wherein the second network model comprises a plurality of third network models resulting from training the first network model; wherein

the processing the grayscale image corresponding to the target sample image by the second network model to obtain the second sample image comprises:

processing the grayscale image by the plurality of third network models, respectively, to obtain a plurality of fifth sample images; and

performing a transparency blending process on the plurality of fifth sample images to obtain the second sample image.

15 . A device of training a metallic texture image generation model, comprising: a processor, and a memory communicatively connected with the processor; wherein

the memory stores computer-executable instructions;

the processor executes the computer-executable instructions stored in the memory to implement the method of claim 8 .

16 . A non-transitory computer-readable storage medium with computer-executable instructions stored thereon, wherein the computer-executable instructions, when executed by a processor, are configured to implement the method of claim 8 .

17 . A device for generating an image with metallic texture, comprising: a processor, and a memory communicatively connected with the processor, wherein

the memory stores computer-executable instructions; wherein the instructions upon execution by the processor, cause the processor to:

acquire a first video; and

input the first video into a pre-trained metallic texture image generation model to obtain a second video, wherein each frame of images in the second video is an image with metallic texture,

wherein the metallic texture image generation model is obtained by updating model parameters of an initial model based on a plurality of sample image pairs,

the plurality of sample image pairs are determined based on a plurality of second sample images and a plurality of target sample images, and

the plurality of second sample images are determined based on a plurality of first sample images with metallic texture, wherein the plurality of first sample images are determined based on the plurality of target sample images, and wherein a difference between a position of a human face in each of the plurality of second sample images and a position of a human face in a corresponding target sample image among the plurality of target sample images is less than a preset value.

18 . The device according to claim 17 , wherein each second sample image is obtained by inputting a grayscale image corresponding to a target sample image into a second network model;

the second network model is obtained by training a first network model based on a third sample image and the grayscale image corresponding to the target sample image; and

the third sample image is determined based on a first sample image, and

wherein the third sample image is based on an alignment of second facial keypoints in the first sample image and first facial keypoints in the target sample image.

19 . The device according to claim 17 , wherein each second sample image is obtained by inputting a grayscale image corresponding to a target sample image into a second network model;

the second network model is obtained by training a first network model based on a third sample image and the grayscale image corresponding to the target sample image; and

the third sample image is determined based on a first sample image, and

wherein the third sample image is based on an alignment of second facial keypoints in the first sample image and first facial keypoints in the target sample image.

20 . The device according to claim 17 , wherein each sample image pair is derived based on a second sample image and a processed image; and

the processed image is obtained by performing an attribute information adjustment process on a target sample image corresponding to the second sample image.