Abstract: An apparatus comprising a processor and a memory that cause the apparatus to perform receiving a video indicating a motion, generating a set of scalar representations of movement based, at least in part, on at least part of the video, and identifying at least one predetermined motion that correlates to the set of scalar representations of movement is disclosed.

Abstract: Techniques are disclosed for indexing and searching high-dimensional data using inverted file structures and product quantization encoding. An image descriptor is quantized using a form of product quantization to determine which of several inverted lists the image descriptor is to be stored. The image descriptor is appended to the corresponding inverted list with a compact coding using a product quantization encoding scheme. When processing a query, a shortlist is computed that includes a set of candidate search results. The shortlist is based on the orthogonality between two random vectors in high-dimensional spaces. The inverted lists are traversed in the order of the distance between the query and the centroid of a coarse quantizer corresponding to each inverted list. The shortlist is ranked according to the distance estimated by a form of product quantization, and the top images referred to by the ranked shortlist are reported as the search results.

Abstract: In embodiments of event image curation, a computing device includes memory that stores a collection of digital images associated with a type of event, such as a digital photo album of digital photos associated with the event, or a video of image frames and the video is associated with the event. A curation application implements a convolutional neural network, which receives the digital images and a designation of the type of event. The convolutional neural network can then determine an importance rating of each digital image within the collection of the digital images based on the type of the event. The importance rating of a digital image is representative of an importance of the digital image to a person in context of the type of the event. The convolutional neural network generates an output of representative digital images from the collection based on the importance rating of each digital image.

Abstract: Techniques for planar region-guided estimates of 3D geometry of objects depicted in a single 2D image. The techniques estimate regions of an image that are part of planar regions (i.e., flat surfaces) and use those planar region estimates to estimate the 3D geometry of the objects in the image. The planar regions and resulting 3D geometry are estimated using only a single 2D image of the objects. Training data from images of other objects is used to train a CNN with a model that is then used to make planar region estimates using a single 2D image. The planar region estimates, in one example, are based on estimates of planarity (surface plane information) and estimates of edges (depth discontinuities and edges between surface planes) that are estimated using models trained using images of other scenes.

Abstract: Techniques for increasing robustness of a convolutional neural network based on training that uses multiple datasets and multiple tasks are described. For example, a computer system trains the convolutional neural network across multiple datasets and multiple tasks. The convolutional neural network is configured for learning features from images and accordingly generating feature vectors. By using multiple datasets and multiple tasks, the robustness of the convolutional neural network is increased. A feature vector of an image is used to apply an image-related operation to the image. For example, the image is classified, indexed, or objects in the image are tagged based on the feature vector. Because the robustness is increased, the accuracy of the generating feature vectors is also increased. Hence, the overall quality of an image service is enhanced, where the image service relies on the image-related operation.

Abstract: In embodiments of image color and tone style transfer, a computing device implements an image style transfer algorithm to generate a modified image from an input image based on a color style and a tone style of a style image. A user can select the input image that includes color features, as well as select the style image that includes an example of the color style and the tone style to transfer to the input image. A chrominance transfer function can then be applied to transfer the color style to the input image, utilizing a covariance of an input image color of the input image to control modification of the input image color. A luminance transfer function can also be applied to transfer the tone style to the input image, utilizing a tone mapping curve based on a non-linear optimization to estimate luminance parameters of the tone mapping curve.

Abstract: A digital medium environment includes an image processing application that performs object segmentation on an input image. An improved object segmentation method implemented by the image processing application comprises receiving an input image that includes an object region to be segmented by a segmentation process, processing the input image to provide a first segmentation that defines the object region, and processing the first segmentation to provide a second segmentation that provides pixel-wise label assignments for the object region. In some implementations, the image processing application performs improved sky segmentation on an input image containing a depiction of a sky.

Abstract: Semantic class localization techniques and systems are described. In one or more implementation, a technique is employed to back communicate relevancies of aggregations back through layers of a neural network. Through use of these relevancies, activation relevancy maps are created that describe relevancy of portions of the image to the classification of the image as corresponding to a semantic class. In this way, the semantic class is localized to portions of the image. This may be performed through communication of positive and not negative relevancies, use of contrastive attention maps to different between semantic classes and even within a same semantic class through use of a self-contrastive technique.

Abstract: In embodiments of convolutional neural network joint training, a computing system memory maintains different data batches of multiple digital image items, where the digital image items of the different data batches have some common features. A convolutional neural network (CNN) receives input of the digital image items of the different data batches, and classifier layers of the CNN are trained to recognize the common features in the digital image items of the different data batches. The recognized common features are input to fully-connected layers of the CNN that distinguish between the recognized common features of the digital image items of the different data batches. A scoring difference is determined between item pairs of the digital image items in a particular one of the different data batches. A piecewise ranking loss algorithm maintains the scoring difference between the item pairs, and the scoring difference is used to train CNN regression functions.

Abstract: In embodiments of event image curation, a computing device includes memory that stores a collection of digital images associated with a type of event, such as a digital photo album of digital photos associated with the event, or a video of image frames and the video is associated with the event. A curation application implements a convolutional neural network, which receives the digital images and a designation of the type of event. The convolutional neural network can then determine an importance rating of each digital image within the collection of the digital images based on the type of the event. The importance rating of a digital image is representative of an importance of the digital image to a person in context of the type of the event. The convolutional neural network generates an output of representative digital images from the collection based on the importance rating of each digital image.

Abstract: Techniques for increasing robustness of a convolutional neural network based on training that uses multiple datasets and multiple tasks are described. For example, a computer system trains the convolutional neural network across multiple datasets and multiple tasks. The convolutional neural network is configured for learning features from images and accordingly generating feature vectors. By using multiple datasets and multiple tasks, the robustness of the convolutional neural network is increased. A feature vector of an image is used to apply an image-related operation to the image. For example, the image is classified, indexed, or objects in the image are tagged based on the feature vector. Because the robustness is increased, the accuracy of the generating feature vectors is also increased. Hence, the overall quality of an image service is enhanced, where the image service relies on the image-related operation.

Abstract: Semantic class localization techniques and systems are described. In one or more implementation, a technique is employed to back communicate relevancies of aggregations back through layers of a neural network. Through use of these relevancies, activation relevancy maps are created that describe relevancy of portions of the image to the classification of the image as corresponding to a semantic class. In this way, the semantic class is localized to portions of the image. This may be performed through communication of positive and not negative relevancies, use of contrastive attention maps to different between semantic classes and even within a same semantic class through use of a self-contrastive technique.

Abstract: Techniques for increasing robustness of a convolutional neural network based on training that uses multiple datasets and multiple tasks are described. For example, a computer system trains the convolutional neural network across multiple datasets and multiple tasks. The convolutional neural network is configured for learning features from images and accordingly generating feature vectors. By using multiple datasets and multiple tasks, the robustness of the convolutional neural network is increased. A feature vector of an image is used to apply an image-related operation to the image. For example, the image is classified, indexed, or objects in the image are tagged based on the feature vector. Because the robustness is increased, the accuracy of the generating feature vectors is also increased. Hence, the overall quality of an image service is enhanced, where the image service relies on the image-related operation.

Abstract: Neural network image curation techniques are described. In one or more implementations, curation is controlled of images that represent a repository of images. A plurality of images of the repository are curated by one or more computing devices to select representative images of the repository. The curation includes calculating a score based on image and face aesthetics, jointly, for each of the plurality of images through processing by a neural network, ranking the plurality of images based on respective said scores, and selecting one or more of the plurality of images as one of the representative images of the repository based on the ranking and a determination that the one or more said images are not visually similar to images that have already been selected as one of the representative images of the repository.

Abstract: Image zooming is described. In one or more implementations, zoomed croppings of an image are scored. The scores calculated for the zoomed croppings are indicative of a zoomed cropping's inclusion of content that is captured in the image. For example, the scores are indicative of a degree to which a zoomed cropping includes salient content of the image, a degree to which the salient content included in the zoomed cropping is centered in the image, and a degree to which the zoomed cropping preserves specified regions-to-keep and excludes specified regions-to-remove. Based on the scores, at least one zoomed cropping may be chosen to effectuate a zooming of the image. Accordingly, the image may be zoomed according to the zoomed cropping such that an amount the image is zoomed corresponds to a scale of the zoomed cropping.

Abstract: Systems and methods are disclosed for estimating aesthetic quality of digital images using deep learning. In particular, the disclosed systems and methods describe training a neural network to generate an aesthetic quality score digital images. In particular, the neural network includes a training structure that compares relative rankings of pairs of training images to accurately predict a relative ranking of a digital image. Additionally, in training the neural network, an image rating system can utilize content-aware and user-aware sampling techniques to identify pairs of training images that have similar content and/or that have been rated by the same or different users. Using content-aware and user-aware sampling techniques, the neural network can be trained to accurately predict aesthetic quality ratings that reflect subjective opinions of most users as well as provide aesthetic scores for digital images that represent the wide spectrum of aesthetic preferences of various users.

Abstract: Embodiments of the present disclosure relate to a sky editing system and related processes for sky editing. The sky editing system includes a composition detector to determine the composition of a target image. A sky search engine in the sky editing system is configured to find a reference image with similar composition with the target image. Subsequently, a sky editor replaces content of the sky in the target image with content of the sky in the reference image. As such, the sky editing system transforms the target image into a new image with a preferred sky background.

Abstract: Systems and methods are disclosed for segmenting a digital image to identify an object portrayed in the digital image from background pixels in the digital image. In particular, in one or more embodiments, the disclosed systems and methods use a first neural network and a second neural network to generate image information used to generate a segmentation mask that corresponds to the object portrayed in the digital image. Specifically, in one or more embodiments, the disclosed systems and methods optimize a fit between a mask boundary of the segmentation mask to edges of the object portrayed in the digital image to accurately segment the object within the digital image.

Abstract: A digital medium environment includes an image processing application that performs object segmentation on an input image. An improved object segmentation method implemented by the image processing application comprises receiving an input image that includes an object region to be segmented by a segmentation process, processing the input image to provide a first segmentation that defines the object region, and processing the first segmentation to provide a second segmentation that provides pixel-wise label assignments for the object region. In some implementations, the image processing application performs improved sky segmentation on an input image containing a depiction of a sky.

Abstract: Embodiments of the present invention provide an automated image tagging system that can predict a set of tags, along with relevance scores, that can be used for keyword-based image retrieval, image tag proposal, and image tag auto-completion based on user input. Initially, during training, a clustering technique is utilized to reduce cluster imbalance in the data that is input into a convolutional neural network (CNN) for training feature data. In embodiments, the clustering technique can also be utilized to compute data point similarity that can be utilized for tag propagation (to tag untagged images). During testing, a diversity based voting framework is utilized to overcome user tagging biases. In some embodiments, bigram re-weighting can down-weight a keyword that is likely to be part of a bigram based on a predicted tag set.