System and method for synthetic image generation with localized editing
Embodiments described herein provide a system for generating synthetic images with localized editing. During operation, the system obtains a source image and a target image for image synthesis and selects a semantic element from the source image. The semantic element indicates a semantically meaningful part of an object depicted in the source image. The system then determines the style information associated with the source and target images. Subsequently, the system generates a synthetic image by transferring the style of the semantic element from the source image to the target image based on the feature representations. In this way, the system can facilitate localized editing of the target image.
1. A method for generating synthetic images with localized editing, comprising:
obtaining a source image and a target image as inputs for image synthesis;
extracting, by applying a first Artificial Intelligence (AI) model, respective feature vectors from the source and target images;
selecting a semantic element from the source image, wherein the semantic element indicates a semantically meaningful part of an object depicted in the source image;
determining respective style information associated with the source and target images;
determining a style of the semantic element from the style information localized at the semantic element; and
generating, using a second AI model, a synthetic image by transferring the style of the selected semantic element from the source image to the target image based on the feature vectors, thereby facilitating the localized editing of the target image.
2. The method of claim 1 , further comprising:
obtaining a strength of the transfer of the style of the semantic element; and
transferring the style of the semantic element based on the strength.
3. The method of claim 1 , wherein the second AI model includes a generative adversarial network (GAN), and wherein the GAN includes a StyleGAN.
4. The method of claim 1 , further comprising:
obtaining corresponding embeddings from one or more layers of the first AI model as the feature vectors associated with the source and target images; and
generating a set of clusters based on the feature vectors, wherein a respective cluster corresponds to a semantic element of the source image.
5. The method of claim 4 , wherein the set of clusters are generated based on one or more of: K-means clustering, spherical K-means clustering, and non-negative matrix factorization.
6. The method of claim 1 , wherein the first AI-model includes a StyleGAN, wherein the source and target images are synthetic images generated by the StyleGAN, and wherein the feature vectors correspond to the embeddings of one or more hidden layers of the StyleGAN.
7. The method of claim 1 , wherein the source and target images are natural images; and
wherein the method further comprises converting the source and target images to respective StyleGAN representations.
8. The method of claim 1 , wherein transferring the semantic element further comprises suppressing transfer of styles outside of the semantic element from the source image.
9. The method of claim 1 , wherein transferring the style of the semantic element further comprises performing style interpolation between the source image and the target image based on a localizing condition matrix associated with the semantic element.
10. The method of claim 1 , further comprising presenting a user interface capable of obtaining a user input that selects the semantic element from the source image, wherein the user interface is configured to obtain the user input based on one or more of: a selection of a spatial location of the semantic element on the source image and a selection from a catalog of semantic elements.
11. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for generating synthetic images with localized editing, the method comprising:
obtaining a source image and a target image as inputs for image synthesis;
extracting, by applying a first Artificial Intelligence (AI) model, respective feature vectors from the source and target images;
selecting a semantic element from the source image, wherein the semantic element indicates a semantically meaningful part of an object depicted in the source image;
determining respective style information associated with the source and target images;
determining a style of the semantic element from the style information localized at the semantic element; and
generating, using a second AI model, a synthetic image by transferring the style of the selected semantic element from the source image to the target image based on the feature vectors, thereby facilitating the localized editing of the target image.
12. The computer-readable storage medium of claim 11 , wherein the method further comprises:
obtaining a strength of the transfer of the style of the semantic element; and
transferring the style of the semantic element based on the strength.
13. The computer-readable storage medium of claim 11 , wherein the second AI model includes a generative adversarial network (GAN), and wherein the GAN includes a StyleGAN.
14. The computer-readable storage medium of claim 11 , wherein the method further comprises:
obtaining corresponding embeddings from one or more layers of the first AI model as the feature vectors associated with the source and target images; and
generating a set of clusters based on the feature vectors, wherein a respective cluster corresponds to a semantic element of the source image.
15. The computer-readable storage medium of claim 14 , wherein the set of clusters are generated based on one or more of: K-means clustering, spherical K-means clustering, and non-negative matrix factorization.
16. The computer-readable storage medium of claim 11 , wherein the first AI-model includes a StyleGAN, wherein the source and target images are synthetic images generated by the StyleGAN, and wherein the feature vectors correspond to the embeddings of one or more hidden layers of the StyleGAN.
17. The computer-readable storage medium of claim 14 , wherein the source and target images are natural images; and
wherein the method further comprises converting the source and target images to respective StyleGAN representations.
18. The computer-readable storage medium of claim 11 , wherein transferring the semantic element further comprises suppressing transfer of styles outside of the semantic element from the source image.
19. The computer-readable storage medium of claim 11 , wherein transferring the style of the semantic element further comprises performing style interpolation between the source image and the target image based on a localizing condition matrix associated with the semantic element.
20. The computer-readable storage medium of claim 11 , wherein the method further comprises presenting a user interface capable of obtaining a user input that selects the semantic element from the source image, wherein the user interface is configured to obtain the user input based on one or more of: a selection of a spatial location of the semantic element on the source image and a selection from a catalog of semantic elements.