Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for debiasing vision-language models utilizing additive residual learning. In particular, in one or more embodiments, the disclosed systems generate an encoded image representation of a digital image utilizing an image encoder of a vision-language neural network. Additionally, in some embodiments, the disclosed systems extract a protected attribute encoding from the encoded image representation of the digital image utilizing an additive residual learner. Upon extracting the protected attribute encoding, in some implementations, the disclosed systems determine a debiased image encoding for the digital image by combining the protected attribute encoding and the encoded image representation.

Abstract: This disclosure describes one or more implementations of systems, non-transitory computer-readable media, and methods that utilize a local implicit image function neural network to perform image segmentation with a continuous class label probability distribution. For example, the disclosed systems utilize a local-implicit-image-function (LIIF) network to learn a mapping from an image to its semantic label space. In some instances, the disclosed systems utilize an image encoder to generate an image vector representation from an image. Subsequently, in one or more implementations, the disclosed systems utilize the image vector representation with a LIIF network decoder that generates a continuous probability distribution in a label space for the image to create a semantic segmentation mask for the image.

Abstract: In implementations of systems for generating digital content, a computing device implements a generation system to receive a user input specifying a characteristic for digital content. The generation system generates input text based on the characteristic for processing by a first machine learning model. Output text generated by the first machine learning model based on processing the input text is received. The output text describes a digital content component. The generation system generates the digital content component by processing the output text using a second machine learning model. The generation system generates the digital content including the digital content component for display in a user interface based on the characteristic.

Abstract: In implementations of systems for generating images for virtual try-on and pose transfer, a computing device implements a generator system to receive input data describing a first digital image that depicts a person in a pose and a second digital image that depicts a garment. Candidate appearance flow maps are computed that warp the garment based on the pose at different pixel-block sizes using a first machine learning model. The generator system generates a warped garment image by combining the candidate appearance flow maps as an aggregate per-pixel displacement map using a convolutional gated recurrent network. A conditional segment mask is predicted that segments portions of a geometry of the person using a second machine learning model. The generator system outputs a digital image that depicts the person in the pose wearing the garment based on the warped garment image and the conditional segmentation mask using a third machine learning model.

Abstract: In implementations of systems for generating images for virtual try-on and pose transfer, a computing device implements a generator system to receive input data describing a first digital image that depicts a person in a pose and a second digital image that depicts a garment. Candidate appearance flow maps are computed that warp the garment based on the pose at different pixel-block sizes using a first machine learning model. The generator system generates a warped garment image by combining the candidate appearance flow maps as an aggregate per-pixel displacement map using a convolutional gated recurrent network. A conditional segment mask is predicted that segments portions of a geometry of the person using a second machine learning model. The generator system outputs a digital image that depicts the person in the pose wearing the garment based on the warped garment image and the conditional segmentation mask using a third machine learning model.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media for one-shot and few-shot image segmentation on classes of objects that were not represented during training. In some embodiments, a dual prediction scheme may be applied in which query and support masks are jointly predicted using a shared decoder, which aids in similarity propagation between the query and support features. Additionally or alternatively, foreground and background attentive fusion may be applied to utilize cues from foreground and background feature similarities between the query and support images. Finally, to prevent overfitting on class-conditional similarities across training classes, input channel averaging may be applied for the query image during training. Accordingly, the techniques described herein may be used to achieve state-of-the-art performance for both one-shot and few-shot segmentation tasks.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: Automated digital asset tagging techniques and systems are described that support use of multiple vocabulary sets. In one example, a plurality of digital assets are obtained having first-vocabulary tags taken from a first-vocabulary set. Second-vocabulary tags taken from a second-vocabulary set are assigned to the plurality of digital assets through machine learning. A determination is made that at least one first-vocabulary tag includes a plurality of visual classes based on the assignment of at least one second-vocabulary tag. Digital assets are collected from the plurality of digital assets that correspond to one visual class of the plurality of visual classes. The model is generated using machine learning based on the collected digital assets.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media for one-shot and few-shot image segmentation on classes of objects that were not represented during training. In some embodiments, a dual prediction scheme may be applied in which query and support masks are jointly predicted using a shared decoder, which aids in similarity propagation between the query and support features. Additionally or alternatively, foreground and background attentive fusion may be applied to utilize cues from foreground and background feature similarities between the query and support images. Finally, to prevent overfitting on class-conditional similarities across training classes, input channel averaging may be applied for the query image during training. Accordingly, the techniques described herein may be used to achieve state-of-the-art performance for both one-shot and few-shot segmentation tasks.

Abstract: Generating a synthesized image of a person wearing clothing is described. A two-dimensional reference image depicting a person wearing an article of clothing and a two-dimensional image of target clothing in which the person is to be depicted as wearing are received. To generate the synthesized image, a warped image of the target clothing is generated via a geometric matching module, which implements a machine learning model trained to recognize similarities between warped and non-warped clothing images using multi-scale patch adversarial loss. The multi-scale patch adversarial loss is determined by sampling patches of different sizes from corresponding locations of warped and non-warped clothing images. The synthesized image is generated on a per-person basis, such that the target clothing fits the particular body shape, pose, and unique characteristics of the person.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating a virtual try-on digital image utilizing a unified neural network framework. For example, the disclosed systems can utilize a coarse-to-fine warping process to generate a warped version of a product digital image to fit a model digital image. In addition, the disclosed systems can utilize a texture transfer process to generate a corrected segmentation mask indicating portions of a model digital image to replace with a warped product digital image. The disclosed systems can further generate a virtual try-on digital image based on a warped product digital image, a model digital image, and a corrected segmentation mask. In some embodiments, the disclosed systems can train one or more neural networks to generate accurate outputs for various stages of generating a virtual try-on digital image.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating a virtual try-on digital image utilizing a unified neural network framework. For example, the disclosed systems can utilize a coarse-to-fine warping process to generate a warped version of a product digital image to fit a model digital image. In addition, the disclosed systems can utilize a texture transfer process to generate a corrected segmentation mask indicating portions of a model digital image to replace with a warped product digital image. The disclosed systems can further generate a virtual try-on digital image based on a warped product digital image, a model digital image, and a corrected segmentation mask. In some embodiments, the disclosed systems can train one or more neural networks to generate accurate outputs for various stages of generating a virtual try-on digital image.

Abstract: Generating a synthesized image of a person wearing clothing is described. A two-dimensional reference image depicting a person wearing an article of clothing and a two-dimensional image of target clothing in which the person is to be depicted as wearing are received. To generate the synthesized image, a warped image of the target clothing is generated via a geometric matching module, which implements a machine learning model trained to recognize similarities between warped and non-warped clothing images using multi-scale patch adversarial loss. The multi-scale patch adversarial loss is determined by sampling patches of different sizes from corresponding locations of warped and non-warped clothing images. The synthesized image is generated on a per-person basis, such that the target clothing fits the particular body shape, pose, and unique characteristics of the person.

Abstract: A computer implemented method and apparatus for preserving structural integrity of 3-D models when printing at varying scales, by use of a cueing model.

Abstract: Systems and methods are disclosed for generating viewpoints and/or digital images of defects in a three-dimensional model. In particular, in one or more embodiments, the disclosed systems and methods generate exterior viewpoints by clustering intersection points between a bounding sphere and rays originating from exterior vertices corresponding to one or more defects. In addition, in one or more embodiments, the disclosed systems and methods generate interior viewpoints by clustering intersection points between one or more medial spheres and rays originating from vertices corresponding to interior vertices corresponding to one or more defects. Furthermore, the disclosed systems and methods can apply colors to vertices corresponding to defects in the three-dimensional model such that adjacent vertices in the three-dimensional model have different colors and are more readily discernable.

Abstract: A computer implemented method and apparatus for embedding a 2D image in a 3D model. The method comprises generating a 3-dimensional (3D) print matrix representing a 2-dimensional (2D) image, wherein the print matrix comprises a plurality of sub-regions, the base plane of each sub-region angled with respect to a top surface of the print matrix so as to produce a plurality of shades, each shade representing a shade of the 2D image; and embedding the print matrix in a (3D) model.

Abstract: Local color information in a 3D mesh is used to enhance fine geometric features such as those in embroidered clothes for 3D printing. In some implementations, vertex color information is used to detect edges and to enhance geometry. In one embodiment, a 3D model is projected into a 2D space to obtain a 2D image, so that pixels that lie on edges in the 2D image can be detected. Further, such edge information is propagated back to the 3D model to enhance the geometry of the 3D model. Other embodiments may be described and/or claimed.