Image capture with artifact remediation
A method includes receiving a digital representation of an image of a subject, applying an image artifact machine learning model to generate a classification of the image as containing an image artifact including an identification of a type of the image artifact, identifying an image remediation process based on the classification, and applying the identified image remediation process to remediate the image artifact.
1 . A method comprising:
receiving a digital representation of an image of a subject;
applying an image artifact machine learning model to generate a classification of the image as containing an image artifact including an identification of a type of the image artifact;
identifying an image remediation process based on the classification by iteratively adjusting camera settings and classifying resulting images following such adjusting of the camera settings; and
applying the identified image remediation process to remediate the image artifact.
2 . The method of claim 1 wherein the image artifact machine learning model comprises a model trained on training images having image artifacts, each training image including a label identifying the type of image artifact.
3 . The method of claim 1 wherein the image is received from a camera and wherein the remediation process comprises adjusting camera settings based on the identification of the type of the image artifact.
4 . The method of claim 3 wherein adjusting the camera settings comprises adjusting a focus setting to an out of focus setting to remediate a Moiré image artifact.
5 . The method of claim 3 herein adjusting the camera settings comprises at least one of adjusting a shutter speed to remediate a black lines image artifact, adjusting a shutter speed and ISO to maintain average brightness and remediate a black lines image artifact.
6 . The method of claim 1 wherein applying the identified image remediation process comprises applying a generative AI model to remediate the artifact and generate a final image.
7 . The method of claim 6 wherein the image artifact machine learning model includes segmentation to identify a segment containing the image artifact and wherein the generative AI model is applied to the segment.
8 . The method of claim 1 wherein identifying the image remediation process comprises:
identifying, based on the identified type of image artifact, one of multiple generative AI models, each model trained on remediating a different type of image artifact; and
using at least one of the multiple generative AI models to generate a final image.
9 . The method of claim 1 wherein the classification the image containing an image artifact identifies an area of the image containing the image artifact, and wherein applying the identified image remediation process comprises:
adjusting camera settings based on the identification of the type of the image artifact;
obtaining a new image based on adjusted camera settings;
applying the image artifact machine learning model to generate a classification of the new image;
detecting that artifacts remain in the new image; and
providing the new image and classification to a generative AI model trained to generate image content for the area to remediate the artifact and generate a final image including the generated image content.
10 . A non-transitory machine-readable storage device having instructions for execution by a processor of a machine to cause the processor to perform operations to perform a method, the operations comprising:
receiving a digital representation of an image of a subject;
applying an image artifact machine learning model to generate a classification of the image as containing an image artifact including an identification of a type of the image artifact;
identifying an image remediation process based on the classification by iteratively adjusting camera settings and classifying resulting images following such adjusting of the camera settings; and
applying the identified image remediation process to remediate the image artifact.
11 . The device of claim 10 wherein the image artifact machine learning model comprises a model trained on training images having image artifacts, each training image including a label identifying the type of image artifact.
12 . The device of claim 10 wherein the image is received from a camera and wherein the remediation process comprises adjusting camera settings based on the identification of the type of the image artifact.
13 . The device of claim 12 wherein adjusting the camera settings comprises adjusting a focus setting to an out of focus setting to remediate a Moiré image artifact.
14 . The device of claim 12 herein adjusting the camera settings comprises at least one of adjusting a shutter speed to remediate a black lines image artifact, adjusting a shutter speed and ISO to maintain average brightness and remediate a black lines image artifact.
15 . The device of claim 10 wherein applying the identified image remediation process comprises applying a generative AI model to remediate the artifact and generate a final image.
16 . The device of claim 10 wherein identifying the image remediation process comprises operations including:
identifying, based on the identified type of image artifact, one of multiple generative AI models, each model trained on remediating a different type of image artifact; and
using at least one of the multiple generative AI models to generate a final image.
17 . A device comprising:
a processor; and
a memory device coupled to the processor and having a program stored thereon for execution by the processor to perform operations comprising:
receiving a digital representation of an image of a subject;
applying an image artifact machine learning model to generate a classification of the image as containing an image artifact including an identification of a type of the image artifact;
identifying an image remediation process based on the classification by iteratively adjusting camera settings and classifying resulting images following such adjusting of the camera settings; and
applying the identified image remediation process to remediate the image artifact.
18 . The device of claim 17 wherein identifying the image remediation process comprises operations including:
identifying, based on the identified type of image artifact, one of multiple generative AI models, each model trained on remediating a different type of image artifact; and
using at least one of the multiple generative AI models to generate a final image.