Systems and methods for automatically adding text content to generated images
In some implementations, the techniques described herein relate to a method including: (i) identifying, by a processor, a generative machine learning model trained on image data, (ii) generating, by the generative machine learning model executed by the processor, an image based on at least one parameter, (iii) editing, by an image-editing algorithm executed by the processor, the image to comprise a specified string of text in a selected area of the image, and (iv) causing display, by the processor, of the edited image.
1 . A method comprising:
identifying, by a processor, a generative machine learning model trained on image data, wherein the generative machine learning model comprises one of a generative adversarial network (GAN) or a diffusion model configured to synthesize new images from training data;
generating, by the generative machine learning model executed by the processor, an image based on at least one parameter, wherein the synthesized image is created by the generative machine learning model without capturing real-world objects;
determining, by the processor, that the image lacks readable text content;
editing, by an image-editing algorithm executed by the processor, the image to comprise a specified string of text in a selected area of the image; and
causing display, by the processor, of the edited image with the specified string of text.
2 . The method of claim 1 , wherein editing, by the image-editing algorithm, the image to comprise the specified string of text comprises:
receiving, by the image-editing algorithm, the specified string of text and the image;
identifying, by the image-editing algorithm, a legible location for the specified string of text within the image; and
adding, by the image-editing algorithm, the specified string of text to the image at the legible location.
3 . The method of claim 2 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font size for the specified string of text.
4 . The method of claim 2 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font color for the specified string of text.
5 . The method of claim 2 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font face for the specified string of text.
6 . The method of claim 2 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises:
identifying, by the image-editing algorithm, at least one visually significant element within the image; and
identifying the legible location that avoids the at least one visually significant element.
7 . The method of claim 6 , wherein identifying the at least one visually significant element comprises using a face-detection algorithm to identify at least one face.
8 . The method of claim 1 , wherein the image data is associated with an organization and causing display of the edited image comprises causing display of the edited image on a platform associated with the organization.
9 . A non-transitory computer-readable storage medium for tangibly storing computer program instructions capable of being executed by a computer processor, the computer program instructions defining steps of:
identifying, by a processor, a generative machine learning model trained on image data, wherein the generative machine learning model comprises one of a generative adversarial network (GAN) or a diffusion model configured to synthesize new images from training data;
generating, by the generative machine learning model executed by the processor, an image based on at least one parameter, wherein the synthesized image is created by the generative machine learning model without capturing real-world objects;
determining, by the processor, that the image lacks readable text content;
editing, by an image-editing algorithm executed by the processor, the image to comprise a specified string of text in a selected area of the image; and
causing display, by the processor, of the edited image.
10 . The non-transitory computer-readable storage medium of claim 9 , wherein editing, by the image-editing algorithm, the image to comprise the specified string of text comprises:
receiving, by the image-editing algorithm, the specified string of text and the image;
identifying, by the image-editing algorithm, a legible location for the specified string of text within the image; and
adding, by the image-editing algorithm, the specified string of text to the image at the legible location.
11 . The non-transitory computer-readable storage medium of claim 10 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font size for the specified string of text.
12 . The non-transitory computer-readable storage medium of claim 10 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font color for the specified string of text.
13 . The non-transitory computer-readable storage medium of claim 10 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font face for the specified string of text.
14 . The non-transitory computer-readable storage medium of claim 10 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises:
identifying, by the image-editing algorithm, at least one visually significant element within the image; and
identifying the legible location that avoids the at least one visually significant element.
15 . The non-transitory computer-readable storage medium of claim 14 , wherein identifying the at least one visually significant element comprises using a face-detection algorithm to identify at least one face.
16 . The non-transitory computer-readable storage medium of claim 9 , wherein the image data is associated with an organization and causing display of the edited image comprises causing display of the edited image on a platform associated with the organization.
17 . A device comprising:
a processor; and
a storage medium for tangibly storing thereon logic for execution by the processor, the logic comprising instructions for:
identifying, by the processor, a generative machine learning model trained on image data associated with an organization, wherein the generative machine learning model comprises one of a generative adversarial network (GAN) or a diffusion model configured to synthesize new images from training data;
generating, by the generative machine learning model executed by the processor, an image based on at least one parameter, wherein the synthesized image is created by the generative machine learning model without capturing real-world objects;
determining, by the processor, that the image lacks readable text content;
editing, by an image-editing algorithm executed by the processor, the image to comprise a specified string of text in a selected area of the image; and
causing display, by the processor, of the edited image on a platform associated with the organization.
18 . The device of claim 17 , wherein editing, by the image-editing algorithm, the image to comprise the specified string of text comprises:
receiving, by the image-editing algorithm, the specified string of text and the image;
identifying, by the image-editing algorithm, a legible location for the specified string of text within the image; and
adding, by the image-editing algorithm, the specified string of text to the image at the legible location.
19 . The device of claim 18 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font size for the specified string of text.
20 . The device of claim 18 , wherein identifying, by the image-editing algorithm, the legible location for the specified string of text within the image comprises selecting, by the image-editing algorithm, a font color for the specified string of text.