Automatic Image Correction Using Machine Learning

A system comprising:

• one or more processors and one or more computer-readable non-transitory storage media coupled to one or more of the processors, the one or more computer-readable non-transitory storage media comprising instructions operable when executed by one or more of the processors to cause the system to perform operations comprising:
• accessing a training image of a person, a reference image of the person, and an incomplete image, wherein the incomplete image corresponds to the training image with a removed portion, wherein the training image, the reference image, and the incomplete image are associated with a training sample of a training dataset;
• generating, using a generator, an in-painted image based on the incomplete image, wherein the in-painted image includes an in-painted portion that corresponds to the removed portion of the incomplete image;
• determining, using a discriminator, whether each of the in-painted image, the training image, and the reference image is likely generated by the generator;
• computing a first loss based on the determination of whether the in-painted image is likely generated by the generator;
• computing a second loss based on the determination of whether the training image is likely generated by the generator;
• computing a third loss based on the determination of whether the reference image is likely generated by the generator;
• updating the discriminator based on the first loss, the second loss, and the third loss;
• determining, using the updated discriminator, whether a second in-painted image generated by the generator is likely generated by the generator, wherein the second in-painted image is associated with a second training image;
• computing a fourth loss based on the determination of whether the second in-painted image is likely generated by the generator;
• and
• updating the generator based on the fourth loss;
• wherein the updated generator is configured to receive an input image and generate a modified version of the input image, wherein the modified version of the input image includes a portion that is in-painted by the updated generator.

One or more computer-readable non-transitory storage media embodying software that is operable when executed to cause one or more processors to perform operations comprising:

• accessing a training image of a person, a reference image of the person, and an incomplete image, wherein the incomplete image corresponds to the training image with a removed portion, wherein the training image, the reference image, and the incomplete image are associated with a training sample of a training dataset;
• generating, using a generator, an in-painted image based on the incomplete image, wherein the in-painted image includes an in-painted portion that corresponds to the removed portion of the incomplete image;
• determining, using a discriminator, whether each of the in-painted image, the training image, and the reference image is likely generated by the generator;
• computing a first loss based on the determination of whether the in-painted image is likely generated by the generator;
• computing a second loss based on the determination of whether the training image is likely generated by the generator;
• computing a third loss based on the determination of whether the reference image is likely generated by the generator;
• updating the discriminator based on the first loss, the second loss, and the third loss;
• determining, using the updated discriminator, whether a second in-painted image generated by the generator is likely generated by the generator, wherein the second in-painted image is associated with a second training image;
• computing a fourth loss based on the determination of whether the second in-painted image is likely generated by the generator;
• and
• updating the generator based on the fourth loss;
• wherein the updated generator is configured to receive an input image and generate a modified version of the input image, wherein the modified version of the input image includes a portion that is in-painted by the updated generator.