EP-3874456-B1 - METHOD AND ELECTRONIC DEVICE FOR PROCESSING FACIAL IMAGES

EP3874456B1EP 3874456 B1EP3874456 B1EP 3874456B1EP-3874456-B1

Inventors

VELUSAMY, SUDHA
VARNA, Rahul
KINI, Ravi Prasad Mohan
REGE, Aniket Devdatta
PARIHAR, Rishubh
HYUN, Daeyoung

Dates

Publication Date: 20260506
Application Date: 20200325

Claims (12)

A method (900) of generating a texture restored image based on a source image, the method comprising: determining information indicative of at least one facial characteristic associated with at least one face in the source image (902), wherein the source image comprises at least one blemish region; generating a filtered image (904) by filtering the source image using a filter based on the determined information indicative of the at least one facial characteristic, wherein the filtered image is a blemish-removed image; determining a number of levels of decomposition to be performed based on the at least one facial characteristic; performing wavelet decomposition for the determined number of levels on each of the filtered image and the source image (906) to generate a first wavelet decomposed image and a second wavelet decomposed image respectively, wherein each level of the first wavelet decomposed image and the second wavelet decomposed image comprises a plurality of wavelet bands; determining weightage factors associated with the wavelet decomposition of each of the filtered image and the source image, based on the determined information indicative of the at least one facial characteristic (908); and obtaining a wavelet image to generate the texture restored image from the wavelet decomposition of each of the filtered image and the source image based on the weightage factors (910), wherein the texture restored image is blemish-removed image that retains underlying skin textures; wherein determining the weightage factors further comprises determining a weightage factor corresponding to each wavelet band at multiple levels of each of the first wavelet decomposed image and the second wavelet decomposed image respectively based on the at least one facial characteristic.
The method (900) of claim 1, further comprising generating an enhanced image by applying image-processing to the texture restored image.
The method (900) of claim 1, wherein the at least one facial characteristic includes at least one from among a level of a skin texture, an amount of blemishes, an average size of blemishes, an age associated with the at least one face, and a gender associated with the at least one face.
The method (900) of claim 1, wherein the generating of the filtered image further comprises: obtaining, from the source image, a region of interest (ROI) indicative of the at least one face in the source image; determining at least one skin region and at least one other region that are associated with the ROI; and generating a masked image based on the at least one skin region and the at least one other region.
The method (900) of claim 4, further comprising: determining at least one area associated with the ROI, the at least one area being indicative of the at least one blemish on the at least one face; generating a blemish-removed image based on the at least one area; and processing the source image using the filter based on the masked image, the blemish-removed image, and the at least one facial characteristic.
The method (900) of claim 1, further comprising: selecting a first set of wavelet bands among multiple bands of the first wavelet decomposed image based on the weightage factor corresponding to each wavelet band of the first wavelet decomposed image; selecting a second set of wavelet bands among multiple bands of the second wavelet decomposed image based on the weightage factor corresponding to each wavelet band of the second wavelet decomposed image; and combining the first set of wavelet bands and the second set of wavelet bands to generate the wavelet image.
An electronic device (100) for generating a texture restored image based on a source image, the electronic device comprising: a processor (106) configured to: determine information indicative of at least one facial characteristic associated with at least one face in the source image, wherein the source image comprises at least one blemish region; generate a filtered image by filtering the source image using a filter based on the determined information indicative of the at least one facial characteristic, wherein the filtered image is a blemish-removed image; determine a number of levels of decomposition to be performed based on the at least one facial characteristic; perform wavelet decomposition for the determined number of levels on each of the filtered image and the source image, respectively to generate a first wavelet decomposed image and a second wavelet decomposed image, wherein each level of the first wavelet decomposed image and the second wavelet decomposed image comprises a plurality of wavelet bands; determine weightage factors associated with the wavelet decomposition of each of the filtered image and the source image, based on the determined information indicative of the at least one facial characteristic; and obtain a wavelet image to generate the texture restored image from the wavelet decomposition of each of the filtered image and the source image based on the weightage factors, wherein the texture restored image is blemish-removed image that retains underlying skin textures; wherein determining the weightage factors further comprises determining a weightage factor corresponding to each wavelet band at multiple levels of each of the first wavelet decomposed image and the second wavelet decomposed image respectively based on the at least one facial characteristic.
The electronic device (100) of claim 7, wherein the processor (106) is further configured to generate an enhanced image by applying image-processing to the texture restored image.
The electronic device (100) of claim 7, wherein the at least one facial characteristic includes at least one from among a level of a skin texture, an amount of blemishes, an average size of blemishes, an age associated with the at least one face, and a gender associated with the at least one face.
The electronic device (100) of claim 7, wherein, in the generating of the filtered image, the processor (106) is further configured to: obtain, from the source image, a region of interest (ROI) indicative of the at least one face in the source image; determine at least one skin region and at least one other region that are associated with the ROI; and generate a masked image based on the at least one skin region and the at least one other region.
The electronic device (100) of claim 10, wherein the processor (106) is further configured to: determine at least one area associated with the ROI, the at least one area being indicative of at least one blemish on the at least one face; generate a blemish-removed image based on the at least one area; and process the source image using the filter based on the masked image, the blemish-removed image, and the at least one facial characteristic.
A non-transitory computer-readable storage medium storing a computer program for operating an electronic device including a memory and a processor, the computer program comprising computer-readable program that, when executed by the processor, causes the electronic device to: determine information indicative of at least one facial characteristic associated with at least one face in a source image, wherein the source image comprises at least one blemish region; generate a filtered image by filtering the source image using a filter based on the determined information indicative of the at least one facial characteristic, wherein the filtered image is a blemish-removed image; determine a number of levels of decomposition to be performed based on the at least one facial characteristic; perform wavelet decomposition for the determined number of levels on each of the filtered image and the source image, respectively to generate a first wavelet decomposed image and a second wavelet decomposed image, wherein each level of the first wavelet decomposed image and the second wavelet decomposed image comprises a plurality of wavelet bands; determine weightage factors associated with the wavelet decomposition of each of the filtered image and the source image, based on the determined information indicative of the at least one facial characteristic; and obtain a wavelet image to generate a texture restored image from the wavelet decomposition of each of the filtered image and the source image based on the weightage factors, wherein the texture restored image is blemish-removed image that retains underlying skin textures; wherein determining the weightage factors further comprises determining a weightage factor corresponding to each wavelet band at multiple levels of each of the first wavelet decomposed image and the second wavelet decomposed image respectively based on the at least one facial characteristic.

Description

Technical Field The disclosure relates to image processing. More particularly, the disclosure relates to a method and an electronic device for processing facial images. Background Art With the advance of technology, the electronic devices such as digital cameras, smart phones, notebooks, laptops, etc., are able to capture and process images to enhance appearance of objects. Faces of people are frequently captured by the image capturing devices, and face enhancement or beautification applications have huge popularity among users, who capture and post self-portrait images on social media. Disclosure Technical Solution In accordance with an aspect of the disclosure, there is provided a method of generating a texture restored image by enhancing a source image. The method includes determining information associated with a source image. The information is indicative of at least one facial characteristic associated with at least one face in the source image. The method includes processing the source image using a filter based on the information associated with the source image. Further, the method includes performing wavelet decomposition on each of the filtered image and the source image. The method also includes determining weightage factors associated with the wavelet decomposition of each of the filtered image and the source image, based on the information associated with the source image. Further, the method includes obtaining a wavelet image to generate the texture restored image from the wavelet decomposition of each of the filtered image and the source image based on the weightage factors for processing the source image. Advantageous Effects The disclosure enables to generate a softened facial image by enhancing the original image. Description of Drawings The above and other aspects, features, and advantages of certain embodiments of the present disclosure will be more apparent from the following description taken in conjunction with the accompanying drawings, in which: FIG. 1 illustrates a block diagram of an electronic device for generating a texture restored image by enhancing a source image, according to an embodiment;FIG. 2A illustrates a flow diagram for a gender determination, according to an embodiment;FIG. 2B illustrates a flow diagram for an age determination, according to an embodiment;FIG. 3 is a flow diagram illustrating a generation of a blemish-removed image, according to an embodiment;FIG. 4 illustrates a flowchart of generating a masked image, according to an embodiment;FIG. 5 illustrates a flow diagram of generating a filtered image, according to an embodiment;FIG. 6 illustrates a block diagram of generating wavelet decomposed images, according to an embodiment;FIG. 7 illustrates a block diagram for weightage factor determination associated with wavelet decomposed images, according to an embodiment;FIG. 8A illustrates a flow diagram of generating a texture restored image, according to an embodiment;FIG. 8B illustrates a flow diagram of generating an enhanced image, according to an embodiment; andFIG. 9 illustrates a flowchart for generating a texture restored image by enhancing a source image, according to an embodiment. Best Mode In accordance with an aspect of the disclosure, there is provided a method of generating a texture restored image by enhancing a source image as set out in accompanying claim 1. decomposition of each of the filtered image and the source image based on the weightage factors for processing the source image. In accordance with an aspect of the disclosure, there is provided an electronic device for generating a texture restored image by enhancing a source image as set out in accompanying claim 7. Mode for Invention The following description with reference to the accompanying drawings is provided to assist in a comprehensive understanding of various embodiments of the disclosure as defined by the claims and their equivalents. It includes various specific details to assist in that understanding but these are to be regarded as merely exemplary. In addition, descriptions of well-known functions and constructions may be omitted for clarity and conciseness of the description. The terms and words used in the following description and claims are not limited to the bibliographical meanings, but, are merely used by the inventor to enable a clear and consistent understanding of the disclosure. Accordingly, it should be apparent to those skilled in the art that the following description of various embodiments of the disclosure is provided for illustration purpose only and not for the purpose of limiting the disclosure as defined by the appended claims and their equivalents. It is to be understood that the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a component surface" includes reference to one or more of such surfaces. As used herein, the terms "1st" or "first" and "2nd" or "second