Abstract: A hybrid search method may be used to identify information responsive to a query. A search may be performed utilizing a neighborhood graph and a partitioning tree. The partitioning tree may be searched to select one or more pivots that may be used to guide a subsequent search in the neighborhood graph. Once the search in the neighborhood graph is unable to identify nearest neighbors in closer proximity to the query, the search may be switched to the partitioning tree. The partitioning tree may then be searched to select pivots that may be used to guide subsequent searches in the neighborhood graph. The searches performed in the partitioning tree and/or the neighborhood graph may be conducted utilizing an iterative algorithm.

Abstract: Some implementations include techniques and arrangements to provide document-related representative information with search results. For example, a representative query and/or representative results may be provided for one or more individual documents identified in a set of search results to supplement the search results returned in response to a received search query. The representative queries may be determined by correlating a plurality of previously submitted queries in search log data with a plurality of documents returned in response to the queries. In some implementations, click-through frequency for a particular document with respect to the plurality of queries may be taken into consideration when determining the representative queries for the particular document.

Abstract: A hybrid search method may be used to identify information responsive to a query. A search may be performed utilizing a neighborhood graph and a partitioning tree. The partitioning tree may be searched to select one or more pivots that may be used to guide a subsequent search in the neighborhood graph. Once the search in the neighborhood graph is unable to identify nearest neighbors in closer proximity to the query, the search may be switched to the partitioning tree. The partitioning tree may then be searched to select pivots that may be used to guide subsequent searches in the neighborhood graph. The searches performed in the partitioning tree and/or the neighborhood graph may be conducted utilizing an iterative algorithm.

Abstract: This disclosure describes various exemplary method and computer program products for transductive multi-label classification in detecting video concepts for information retrieval. This disclosure describes utilizing a hidden Markov random field formulation to detect labels for concepts in a video content and modeling a multi-label interdependence between the labels by a pairwise Markov random field. The process groups the labels into several parts to speed up a labeling inference and calculates a conditional probability score for the labels, the conditional probability scores are ordered for ranking in a video retrieval evaluation.

Abstract: Visual features of images are translated into visual words defined by a dictionary. The visual words are indexed and the images are stored in an image store. A sketched imaged, translated into visual words, is utilized to search for similar images in the image store. The visual words are compared to visual words in the index to identify matches associated with stored images. The stored images are displayed and ranked according to the highest number of matches. Textual searches are used to supplement or refine the search results.

Abstract: A general framework for video search reranking is disclosed which explicitly formulates reranking into a global optimization problem from the Bayesian perspective. Under this framework, with two novel pair-wise ranking distances, two effective video search reranking methods, hinge reranking and preference strength reranking, are disclosed. Experiments conducted on the TRECVID dataset have demonstrated that the disclosed methods outperform several existing reranking approaches.

Abstract: Techniques for constructing an optimized kd-tree are described. In an implementation, an optimized kd-tree process receives input of a set of data points applicable for large-scale computer vision applications. The process divides the set of the data points into subsets of data points with nodes while generating hyperplanes (e.g., coordinate axes). The process identifies a partition axis for each node based on the coordinate axes combined in a binary way. The optimized kd-tree process creates an optimized kd-tree that organizes the data points based on the identified partition axis. The organization of the data points in the optimized kd-tree provides efficient indexing and searching for a nearest neighbor.

Abstract: Region-based image manipulation can include selecting and segmenting regions of a particular image. The regions are identified through the use of simplified brushstrokes over pixels of the regions. Identified regions can be manipulated or transformed accordingly. Certain implementations include filling in regions with other images or objects, and include performing a text query to search for such images or objects.

Abstract: Techniques for intelligent image search results summarization and browsing scheme are described. Images having visual attributes are evaluated for similarities based in part on their visual attributes. At least one preference score indicating a probability of an image to be selected into a summary is calculated for each image. Images are selected based on the similarity of the selected images to the other images and the preference scores of the selected images. A summary of the plurality of images is generated including the selected one individual image.

Abstract: Techniques for image search using contextual information related to a user query are described. A user query including at least one of textual data or image data from a collection of data displayed by a computing device is received from a user. At least one other subset of data selected from the collection of data is received as contextual information that is related to and different from the user query. Data files such as image files are retrieved and ranked based on the user query to provide a pre-ranked set of data files. The pre-ranked data files are then ranked based on the contextual information to provide a re-ranked set of data files to be displayed to the user.

Abstract: The present disclosure describes a user interface and techniques to use the user interface to specify user search intent for color layout image search. In one aspect, a user interface is comprised of a color map control, a plurality of controls to draw a color layout, and a drawing canvas upon which to draw. The controls used to draw the color layout include shape preserving and shape modifying operations as wells as editing operations. In this way a color layout may be automatically generated with drawing semantics. The user interface may also be integrated with text search engines that perform text search on text metadata tags associated with digital images.

Abstract: The concept-structured image search technique described herein pertains to a technique for enabling a user to indicate their semantic intention and then retrieve and rank images from a database or other image set according to this intention. The concept-structured image search technique described herein includes a new interface for image search. With this interface, a user can freely type several key textual words in arbitrary positions on a blank image, and also describe a region for each keyword that indicates its influence scope, which is called concept structure herein. The concept-structured image search technique will return and rank images that are in accordance with the concept structure indicated by the user. One embodiment of the technique can be used to create a synthesized image without actually using the synthesized image to perform a search of an image set.

Abstract: Described is a technology for computing visual and textual summaries for tagged image collections. Heterogeneous affinity propagation is used to together identify both visual and textual exemplars. The heterogeneous affinity propagation finds the exemplars for relational heterogeneous data (e.g., images and words) by considering the relationships (e.g., similarities) within pairs of images, pairs of words, and relationships of words to images (affinity) in an integrated manner.

Abstract: This disclosure describes various exemplary user interfaces, methods, and computer program products for the interactively ranking image search results refinement method using a color layout. The method includes receiving a text query for an image search, presenting image search results in a structured presentation based on the text query and information from an interest color layout. The process creates image search results that may be selected by the user based on color selection palettes or color layout specification schemes. Then the process ranks the image search results by sorting the results according to similarity scores between color layouts from the image search results and the interest color layout from a user based on the color selection palettes and the color layout specification schemes.

Abstract: This disclosure describes various exemplary method and computer program products for transductive multi-label classification in detecting video concepts for information retrieval. This disclosure describes utilizing a hidden Markov random field formulation to detect labels for concepts in a video content and modeling a multi-label interdependence between the labels by a pairwise Markov random field. The process groups the labels into several parts to speed up a labeling inference and calculates a conditional probability score for the labels, the conditional probability scores are ordered for ranking in a video retrieval evaluation.

Abstract: A general framework for video search reranking is disclosed which explicitly formulates reranking into a global optimization problem from the Bayesian perspective. Under this framework, with two novel pair-wise ranking distances, two effective video search reranking methods, hinge reranking and preference strength reranking, are disclosed. Experiments conducted on the TRECVID dataset have demonstrated that the disclosed methods outperform several existing reranking approaches.