IP Library Granted Patent US 11,657,481
Granted Patent B2
US 11,657,481 · App. 16/951,929 · Granted May 23, 2023

Systems and methods for selective enhancement of skin features in images

Inventors: Dmitry Sytnik (Kyiv, UA); Andrey Frolov (Kyiv Oblast, UA)
Assignee: Shinyfields Limited
G06T5/005G06T5/002G06T5/50G06T7/0012G06T2207/20084G06T2207/20221G06T2207/30088G06T2207/30201
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,657,481
App. No.
16/951,929
Granted
May 23, 2023
Kind
B2
Abstract

Exemplary embodiments are directed to a system for selective enhancement of skin features in an image. The system includes an interface configured to receive as input an original image, and a processing device in communication with the interface. The processing device is configured to process the original image using a neural network to detect one or more skin imperfections in the original image, generate a neural network mask of the original image for the one or more skin imperfections in the original image, generate one or more source patches based on the original image, and, replace the one or more skin imperfections in the original image with the one or more source patches to generate a patched skin image.

Claims (34)

1. A system for selective enhancement of skin features in an image, the system comprising:

an interface configured to receive as input an original image; and

a processing device in communication with the interface, the processing device configured to:

(i) process the original image using a neural network to detect one or more skin imperfections in the original image;

(ii) generate a neural network mask of the original image for the one or more skin imperfections in the original image;

(iii) generate one or more source patches based on the original image; and

(iv) replace the one or more skin imperfections in the original image with the one or more source patches to generate a patched skin image;

wherein the processing device is further configured to:

generate a masked skin image including a skin mask; and

generate a blurred image, the blurred image including blurring of the skin encompassed by the skin mask without affecting facial feature details.

2. The system of claim 1 , wherein the original image includes at least one individual with the one or more skin imperfections on a face of the individual.

3. The system of claim 1 , wherein the processing device generates a bounding box around detected skin features in the original image for enhancement, the skin features including the one or more skin imperfections.

4. The system of claim 3 , wherein the processing device generates a separate bounding box for each individual depicted in the original image.

5. The system of claim 1 , wherein the neural network mask is a skin imperfections mask, the skin imperfections mask including an island disposed over and associated with each of the one or more skin imperfections.

6. The system of claim 1 , wherein the processing device generates a defect area independently surrounding each of the one or more skin imperfections.

7. The system of claim 6 , wherein the processing device selects one of the one or more source patches for replacement of one of the one or more skin imperfections based on at least a partial overlap between the defect area and the source patch.

8. The system of claim 1 , wherein the skin mask encompasses skin within the patched skin image and excludes facial feature details from the skin mask.

9. The system of claim 8 , wherein the facial feature details include at least one of eyebrows, hair, nose, or lips.

10. The system of claim 1 , wherein the processing device generates a detail image, the detail image including facial feature details excluded from the skin mask.

11. The system of claim 10 , wherein the processing device generates two or more filtered images.

12. The system of claim 11 , wherein the two or more filtered images include the facial feature details at different kernel sizes, wherein the different kernel sizes are small kernels, medium kernels, and big kernels.

13. The system of claim 12 , wherein the processing device generates a combined image, the combined image including the facial feature details of the small kernels and including only some of the facial feature details of the medium and big kernels.

14. The system of claim 13 , wherein the processing device generates a dark circle mask for shadowed features under eyes of the individual.

15. The system of claim 13 , wherein the processing device generates a noise image, the noise image including a noise effect applied to skin of an individual with the one or more skin imperfections.

16. A system for selective enhancement of skin features in an image, the system comprising:

an interface configured to receive as input an original image; and

a processing device in communication with the interface, the processing device configured to:

(i) process the original image using a neural network to detect one or more skin imperfections in the original image;

(ii) generate a neural network mask of the original image for the one or more skin imperfections in the original image;

(iii) generate one or more source patches based on the original image; and

(iv) replace the one or more skin imperfections in the original image with the one or more source patches to generate a patched skin image;

wherein:

the interface includes an image selection section with the patched skin image and one or more additional original images; and

the interface includes a first submenu for selecting the patched skin image and copying one or more enhancements applied to the patched skin image, and the interface includes a second submenu for selecting one or more of the additional original images and applying the copied one or more enhancements of the patched skin image to the selected one or more of the additional original images.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 18, 2020
From: SYTNIK, DMITRY; FROLOV, ANDREY
To: SHINYFIELDS LIMITED
Reel/Frame 054412/0299 →
Continuity (2)
Provisional Application 62936862 · Nov 18, 2019
Related Publication 20210150677A1 · May 20, 2021
Cited By (1)
US 12,670,542