Abstract: This disclosure describes a system, method, and computer storage medium for joint asset and defect detection. The approach includes receiving input data including an input image of a utility asset, the input image including one or more objects. Deep neural networks are configured to generate embeddings for classification labels of the one or more objects, each embedding corresponding to a classification label and including a mapping between the classification label and a subset of feature vectors. Defect classifiers are configured to determine a likelihood of an object from the one or more objects in the input image containing a type of defect. Each defect classifier is trained to determine a type of defect based on the embeddings for the one or more classification labels. The approach includes generating an output image that includes bounding boxes for the objects and an annotation corresponding a respective object from the objects.

Abstract: This disclosure describes a system, method, and non-transitory computer-readable medium for image search-based object detection of utility assets in image databases. The method includes receiving an input image of a utility asset and a query bounding box representing an image-based object query. Bounding boxes of objects represented in the input image are generated based on the input image and the query bounding box, in which anchoring boxes corresponding to object classifications are identified from the bounding boxes. A textual label is determined for a selected subset of anchoring boxes. An image embedding representing the region is encoded, and image tokens are generated based on the encoded image embedding. Output images of other utility assets relevant to the image-based object query are identified from images in an image database, based on at least one of (i) the encoded image embedding, (ii) the image tokens, or (iii) the textual label. The output images are provided for output.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a storage device, for mapping an electrical grid are disclosed. A method includes sampling multiple locations within a geographic region, executing a detection process for each location, the detection process including applying the set of images for the location as input to a machine learning (ML) model that is trained to identify electrical grid assets depicted within images taken from a combination of different perspectives and obtaining an output from the machine learning model that indicates whether a same electrical grid asset is identified in each of the images of the location. In response to an ML output that indicates a positive identification of the same electrical grid asset being depicted in a particular set of images of a particular location, the method further includes: selecting a number of sublocations within the region, and executing the detection process for each sublocation.

Abstract: Many users make use of search engines to locate desired internet content by submitting search queries. For example, a user may search for photos, applications, websites, videos, documents, and/or information regarding people, places, and things. Unfortunately, search engines may provide a plethora of information that a user may be left to sift through to find relevant content. Accordingly, one or more systems and/or techniques for organizing search results are disclosed herein. In particular, user generated content, such as photos, may be retrieved based upon a search query. The user generated content may be grouped into clusters of user generated content having similar features. Search results of the search query may be obtained and organized based upon comparing the search results with the clusters. The organized search results and/or a table of content including the clusters may be presented to provide an enhanced user experience.

Abstract: Many users make use of search engines to locate desired internet content by submitting search queries. For example, a user may search for photos, applications, websites, videos, documents, and/or information regarding people, places, and things. Unfortunately, search engines may provide a plethora of information that a user may be left to sift through to find relevant content. Accordingly, one or more systems and/or techniques for organizing search results are disclosed herein. In particular, user generated content, such as photos, may be retrieved based upon a search query. The user generated content may be grouped into clusters of user generated content having similar features. Search results of the search query may be obtained and organized based upon comparing the search results with the clusters. The organized search results and/or a table of content comprising the clusters may be presented to provide an enhanced user experience.

Abstract: A candidate for selection or insertion may be based on extracted candidate features and reference features. The candidate features may be scored and ranked while candidates may be removed from consideration based on criteria. The remaining candidates may be presented for selection and/or insertion.

Abstract: Techniques for identifying near-duplicates of a media object and associating metadata of the near-duplicates with the media object are described herein. One or more devices implementing the techniques are configured to identify the near duplicates based at least on similarity attributes included in the media object. Metadata is then extracted from the near-duplicates and is associated with the media object as descriptors of the media object to enable discovery of the media object based on the descriptors.

Abstract: Many users make use of search engines to locate desired internet content by submitting search queries. For example, a user may search for photos, applications, websites, videos, documents, and/or information regarding people, places, and things. Unfortunately, search engines may provide a plethora of information that a user may be left to sift through to find relevant content. Accordingly, one or more systems and/or techniques for organizing search results are disclosed herein. In particular, user generated content, such as photos, may be retrieved based upon a search query. The user generated content may be grouped into clusters of user generated content having similar features. Search results of the search query may be obtained and organized based upon comparing the search results with the clusters. The organized search results and/or a table of content based upon the clusters may be presented to provide an enhanced user experience.

Abstract: Techniques are described for online real time text to image translation suitable for virtually any submitted query. Semantic classes and associated analogous items for each of the semantic classes are determined for the submitted query. One or more requests are formulated that are associated with analogous items. The requests are used to obtain web based images and associated surrounding text. The web based images are used to obtain associated near-duplicate images. The surrounding text of images is analyzed to create high-quality text associated with each semantic class of the submitted query. One or more query dependent classifiers are trained online in real time to remove noisy images. A scoring function is used to score the images. The images with the highest score are returned as a query response.

Abstract: An annotation suggestion platform may comprise a client and a server, where the client captures a media object and sends the captured object to the server, and the server provides a list of suggested annotations for a user to associate with the captured media object. The user may then select which of the suggested metadata is to be associated or stored with the captured media. In this way, a user may more easily associate metadata with a media object, facilitating the media object's search and retrieval. The server may also provide web page links related to the captured media object. Suggested metadata may be viewed, selected, and associated through a user interface. The user interface may additionally provide space for web pages to be navigated and viewed.

Abstract: A candidate for selection or insertion may be based on extracted candidate features and reference features. The candidate features may be scored and ranked while candidates may be removed from consideration based on criteria. The remaining candidates may be presented for selection and/or insertion.

Abstract: Implementations for identifying duplicate images in an image space are described. An image space is partitioned into a plurality of coarse clusters based on signatures of the images within the image space. The signatures are determined from compact descriptors of the images. Refined clusters that include one or more images of an individual coarse cluster are created based on pair-wise comparisons of the compact descriptors of images in the coarse cluster, and the refined clusters are identified as sets of duplicate images. The refined clusters are grown by searching in similar coarse clusters for images to add to the refined clusters.

Abstract: An annotation suggestion platform is described herein. The annotation suggestion platform may comprise a client and a server, where the client captures a media object and sends the captured object to the server, and the server provides a list of suggested annotations for a user to associate with the captured media object. The user may then select which of the suggested metadata is to be associated or stored with the captured media. In this way, a user may more easily associate metadata with a media object, facilitating the media object's search and retrieval. The server may also provide web page links related to the captured media object. A user interface for the annotation suggestion platform is also described herein, as are optimizations including indexing and tag propagation.

Abstract: An annotation suggestion platform may comprise a client and a server, where the client captures a media object and sends the captured object to the server, and the server provides a list of suggested annotations for a user to associate with the captured media object. The user may then select which of the suggested metadata is to be associated or stored with the captured media. In this way, a user may more easily associate metadata with a media object, facilitating the media object's search and retrieval. The server may also provide web page links related to the captured media object. A user interface for the annotation suggestion platform is also described herein, as are optimizations including indexing and tag propagation.

Abstract: While browsing, a user may interact with a wide variety of images. The user may upload and share images taken with a digital camera and/or search for image using a search engine. Because images are rich in contextual information, it may be advantageous to provide additional information, such as adjacent market advertising based upon matching advertisements with contextual information of the images. Accordingly, a query image may be used to retrieve a video frame set. The video frame set may be expanded with related video frames corresponding to adjacent markets. The expanded video frame set may be grouped into clusters of similar frames. The clusters may be used to rank advertisements based upon how similar the advertisements are to the clusters and/or video frames within the clusters. In this way, one or more ranked advertisements may be presented with the query image.

Abstract: Search queries for images are received from users. An original order of responsive images to the query is determined. Duplicate images and words associated with the duplicate images are identified for each of the responsive images. Common words associated with the duplicate images are identified. The responsive images are annotated with the common words and an annotated order is determined. A re-ranked order is determined based on the original order and the annotated order. Responsive images are presented to the user in the re-ranked order.

Abstract: Implementations for identifying duplicate images in an image space are described. An image space is partitioned into a plurality of coarse clusters based on signatures of the images within the image space. The signatures are determined from compact descriptors of the images. Refined clusters that include one or more images of an individual coarse cluster are created based on pair-wise comparisons of the compact descriptors of images in the coarse cluster, and the refined clusters are identified as sets of duplicate images. The refined clusters are grown by searching in similar coarse clusters for images to add to the refined clusters.

Abstract: Names of entities, such as people, in an image may be identified automatically. Visually similar images of entities are retrieved, including text proximate to the visually similar images. The collected text is mined for names of entities, and the detected names are analyzed. A name may be associated with the entity in the image, based on the analysis.

Abstract: Search queries for images are received from users. An original order of responsive images to the query is determined. Duplicate images and words associated with the duplicate images are identified for each of the responsive images. Common words associated with the duplicate images are identified. The responsive images are annotated with the common words and an annotated order is determined. A re-ranked order is determined based on the original order and the annotated order. Responsive images are presented to the user in the re-ranked order.

Abstract: In some implementations, a plurality of first questions and corresponding first answers are identified at a community question-answer (CQA) site as a plurality of first question-answer (q-a) pairs. A query thread comprised of a second question and a plurality of candidate second answers is selected for making a determination of answer quality. A set of the first questions that are similar to the second question are identified from the plurality of first questions. First linking features between the identified set of first questions and their corresponding first answers are used for determining an analogy with second linking features between the second question and candidate answers for ranking the candidate answers.