Abstract: Techniques and systems are described to model and extract knowledge from images. A digital medium environment is configured to learn and use a model to compute a descriptive summarization of an input image automatically and without user intervention. Training data is obtained to train a model using machine learning in order to generate a structured image representation that serves as the descriptive summarization of an input image. The images and associated text are processed to extract structured semantic knowledge from the text, which is then associated with the images. The structured semantic knowledge is processed along with corresponding images to train a model using machine learning such that the model describes a relationship between text features within the structured semantic knowledge. Once the model is learned, the model is usable to process input images to generate a structured image representation of the image.

Abstract: Image cropping suggestion using multiple saliency maps is described. In one or more implementations, component scores, indicative of visual characteristics established for visually-pleasing croppings, are computed for candidate image croppings using multiple different saliency maps. The visual characteristics on which a candidate image cropping is scored may be indicative of its composition quality, an extent to which it preserves content appearing in the scene, and a simplicity of its boundary. Based on the component scores, the croppings may be ranked with regard to each of the visual characteristics. The rankings may be used to cluster the candidate croppings into groups of similar croppings, such that croppings in a group are different by less than a threshold amount and croppings in different groups are different by at least the threshold amount. Based on the clustering, croppings may then be chosen, e.g., to present them to a user for selection.

Abstract: Techniques and systems are described to recolor a group of images for color consistency. Techniques include extracting color palettes for images of the group of images and generating a group theme color palette based on the color palettes for the images. Image color palettes are then mapped to the group theme color palette and the images are modified in response to the mapping. In some examples, the mapping includes discouraging multiple colors of a single color palette from mapping to a single color of the group theme color palette. Additionally, or alternatively, the mapping includes discouraging a forced mapping of a dissimilar color of an image color palette from mapping to the group theme color palette.

Abstract: Stereo correspondence and depth sensor techniques are described. In one or more implementations, a depth map generated by a depth sensor is leveraged as part of processing of stereo images to assist in identifying which parts of stereo images correspond to each other. The depth map, for instance, may be utilized to assist in identifying depth discontinuities in the stereo images. Additionally, techniques may be employed to align the depth discontinuities identified from the depth map to image edges identified from the stereo images. Techniques may also be employed to suppress image edges that do not correspond to the depth discontinuities of the depth map in comparison with image edges that do correspond to the depth discontinuities as part of the identification.

Abstract: Combined selection tool techniques are described in which selection of portions within an image is enabled via a tool configured to selectively switch between a coarse selection mode and a refinement selection mode. In one or more implementations, an image is exposed for editing in a user interface and input is obtained to select portions of the image using the tool. The selection mode is automatically toggled between the coarse selection mode and refinement selection mode based on characteristics of the input, such as position and velocity of the input or gestures mapped to the input. Then, selection boundaries defining the selection of portions of the image are modified in accordance with the input. In one approach, the coarse selection mode corresponds to a graph cut mechanism and the refinement selection mode corresponds to a level set mechanism.

Abstract: Techniques and systems are described for style-aware patching of a digital image in a digital medium environment. For example, a digital image creation system generates style data for a portion to be filled of a digital image, indicating a style of an area surrounding the portion. The digital image creation system also generates content data for the portion indicating content of the digital image of the area surrounding the portion. The digital image creation system selects a source digital image based on similarity of both style and content of the source digital image at a location of the patch to the style data and content data. The digital image creation system transforms the style of the source digital image based on the style data and generates the patch from the source digital image in the transformed style for incorporation into the portion to be filled of the digital image.

Abstract: Stereo correspondence and depth sensor techniques are described. In one or more implementations, a depth map generated by a depth sensor is leveraged as part of processing of stereo images to assist in identifying which parts of stereo images correspond to each other. The depth map, for instance, may be utilized to assist in identifying depth discontinuities in the stereo images. Additionally, techniques may be employed to align the depth discontinuities identified from the depth map to image edges identified from the stereo images. Techniques may also be employed to suppress image edges that do not correspond to the depth discontinuities of the depth map in comparison with image edges that do correspond to the depth discontinuities as part of the identification.

Abstract: Techniques and systems are described to support digital image processing through use of an image repository, e.g., a stock image database or other storage. In one example, a plurality of candidate digital images are obtained from an image repository based on a target digital image. A plurality of transformations are generated to be applied to the target digital image, each transformation based on a respective candidate digital image. Semantic information is employed as part of the transformations, e.g., blending, filtering, or alignment. A plurality of transformed target digital images are generated based at least in part through application of the plurality of transformations to the target image.

Abstract: Techniques and systems are described to model and extract knowledge from images. A digital medium environment is configured to learn and use a model to compute a descriptive summarization of an input image automatically and without user intervention. Training data is obtained to train a model using machine learning in order to generate a structured image representation that serves as the descriptive summarization of an input image. The images and associated text are processed to extract structured semantic knowledge from the text, which is then associated with the images. The structured semantic knowledge is processed along with corresponding images to train a model using machine learning such that the model describes a relationship between text features within the structured semantic knowledge. Once the model is learned, the model is usable to process input images to generate a structured image representation of the image.

Abstract: Stereo correspondence and depth sensor techniques are described. In one or more implementations, a depth map generated by a depth sensor is leveraged as part of processing of stereo images to assist in identifying which parts of stereo images correspond to each other. The depth map, for instance, may be utilized to assist in identifying depth discontinuities in the stereo images. Additionally, techniques may be employed to align the depth discontinuities identified from the depth map to image edges identified from the stereo images. Techniques may also be employed to suppress image edges that do not correspond to the depth discontinuities of the depth map in comparison with image edges that do correspond to the depth discontinuities as part of the identification.

Abstract: Techniques and systems are described to recolor a group of images for color consistency. Techniques include extracting color palettes for images of the group of images and generating a group theme color palette based on the color palettes for the images. Image color palettes are then mapped to the group theme color palette and the images are modified in response to the mapping. In some examples, the mapping includes discouraging multiple colors of a single color palette from mapping to a single color of the group theme color palette. Additionally, or alternatively, the mapping includes discouraging a forced mapping of a dissimilar color of an image color palette from mapping to the group theme color palette.

Abstract: Image cropping suggestion using multiple saliency maps is described. In one or more implementations, component scores, indicative of visual characteristics established for visually-pleasing croppings, are computed for candidate image croppings using multiple different saliency maps. The visual characteristics on which a candidate image cropping is scored may be indicative of its composition quality, an extent to which it preserves content appearing in the scene, and a simplicity of its boundary. Based on the component scores, the croppings may be ranked with regard to each of the visual characteristics. The rankings may be used to cluster the candidate croppings into groups of similar croppings, such that croppings in a group are different by less than a threshold amount and croppings in different groups are different by at least the threshold amount. Based on the clustering, croppings may then be chosen, e.g., to present them to a user for selection.

Abstract: Techniques and systems are described to recolor a group of images for color consistency. Techniques include extracting color palettes for images of the group of images and generating a group theme color palette based on the color palettes for the images. Image color palettes are then mapped to the group theme color palette and the images are modified in response to the mapping. In some examples, the mapping includes discouraging multiple colors of a single color palette from mapping to a single color of the group theme color palette. Additionally, or alternatively, the mapping includes discouraging a forced mapping of a dissimilar color of an image color palette from mapping to the group theme color palette.

Abstract: Image cropping suggestion using multiple saliency maps is described. In one or more implementations, component scores, indicative of visual characteristics established for visually-pleasing croppings, are computed for candidate image croppings using multiple different saliency maps. The visual characteristics on which a candidate image cropping is scored may be indicative of its composition quality, an extent to which it preserves content appearing in the scene, and a simplicity of its boundary. Based on the component scores, the croppings may be ranked with regard to each of the visual characteristics. The rankings may be used to cluster the candidate croppings into groups of similar croppings, such that croppings in a group are different by less than a threshold amount and croppings in different groups are different by at least the threshold amount. Based on the clustering, croppings may then be chosen, e.g., to present them to a user for selection.

Abstract: Image depth inference techniques and systems from semantic labels are described. In one or more implementations, a digital medium environment includes one or more computing devices to control a determination of depth within an image. Regions of the image are semantically labeled by the one or more computing devices. At least one of the semantically labeled regions is decomposed into a plurality of segments formed as planes generally perpendicular to a ground plane of the image. Depth of one or more of the plurality of segments is then inferred based on relationships of respective segments with respective locations of the ground plane of the image. A depth map is formed that describes depth for the at least one semantically labeled region based at least in part on the inferred depths for the one or more of the plurality of segments.

Abstract: Techniques and systems are described to support digital image processing through use of an image repository, e.g., a stock image database or other storage. In one example, a plurality of candidate digital images are obtained from an image repository based on a target digital image. A plurality of transformations are generated to be applied to the target digital image, each transformation based on a respective candidate digital image. Semantic information is employed as part of the transformations, e.g., blending, filtering, or alignment. A plurality of transformed target digital images are generated based at least in part through application of the plurality of transformations to the target image.

Abstract: Stereo correspondence model fitting techniques are described. In one or more implementations, a model may be fit to a region in at least one of a plurality of stereoscopic images of an image scene. The model may then be used as part of a stereo correspondence calculation, which may include computing disparities for the region based at least in part on correspondence to the model.

Abstract: Probabilistic determination of selected image portions is described. In one or more implementations, a selection input is received for selecting a portion of an image. For pixels of the image that correspond to the selection input, probabilities are determined that the pixels are intended to be included as part of a selected portion of the image. In particular, the probability that a given pixel is intended to be included as part of the selected portion of the image is determined as a function of position relative to center pixels of the selection input as well as a difference in one or more visual characteristics with the center pixels. The determined probabilities can then be used to segment the selected portion of the image from a remainder of the image. Based on the segmentation of the selected portion from the remainder of the image, selected portion data can be generated that defines the selected portion of the image.

Abstract: Techniques and systems are described to support digital image processing through use of an image repository, e.g., a stock image database or other storage. In one example, a plurality of candidate digital images are obtained from an image repository based on a target digital image. A plurality of transformations are generated to be applied to the target digital image, each transformation based on a respective candidate digital image. Semantic information is employed as part of the transformations, e.g., blending, filtering, or alignment. A plurality of transformed target digital images are generated based at least in part through application of the plurality of transformations to the target image.

Abstract: Techniques and systems are described for style-aware patching of a digital image in a digital medium environment. For example, a digital image creation system generates style data for a portion to be filled of a digital image, indicating a style of an area surrounding the portion. The digital image creation system also generates content data for the portion indicating content of the digital image of the area surrounding the portion. The digital image creation system selects a source digital image based on similarity of both style and content of the source digital image at a location of the patch to the style data and content data. The digital image creation system transforms the style of the source digital image based on the style data and generates the patch from the source digital image in the transformed style for incorporation into the portion to be filled of the digital image.