Abstract: Compression of an image is performed based on prediction of target blocks of an image from candidate source blocks of the image. Heuristics are used for identifying the candidate source blocks, for example, source blocks are selected from within a cluster of similar blocks obtained by K-means clustering. For each target block, a region adjacent to the target block is identified and a set of candidate source blocks along with candidate source regions adjacent to the candidate source blocks are identified. The candidate source regions are ranked based on the differences between the candidate source regions and the target source region. Each candidate source block is described using its rank and residual information describing differences between the candidate source block and the target block. The candidate source block that can be described using a minimum amount of information is selected for predicting the target block.

Abstract: An image processing system is provided for encoding images based on example-based compression. The system selects a set of candidate dictionary predictor entries to encode a portion of an image based at least in part on the neighbors of the portion. The spatial continuity between portions of the image is exploited by the image processing system by selecting corresponding dictionary predictor entries that have the same offset vector as the portion of the image and its neighboring portions.

Abstract: An exemplar dictionary is built from example image blocks for determining predictor blocks for encoding and decoding images. The exemplar dictionary comprises a hierarchical organization of example image blocks. The hierarchical organization of image blocks is obtained by clustering a set of example image blocks, for example, based on k-means clustering. Performance of clustering is improved by transforming feature vectors representing the image blocks to fewer dimensions. Principal component analysis is used for determining feature vectors with fewer dimensions. The clustering performed at higher levels of the hierarchy uses fewer dimensions of feature vectors compared to lower levels of hierarchy. Performance of clustering is improved by processing only a sample of the image blocks of a cluster. The clustering performed at higher levels of the hierarchy uses lower sampling rates as compared to lower levels of hierarchy.

Abstract: Systems and methods of evaluating information via a computer network are provided. A content group can be identified, and each item of the content group can be associated with a vector indicating at least one user interest category of users exposed to the item. The vectors of each item can be evaluated to generate a first nearest neighbor list of each item of the content group. The nearest neighbor list of a first item can be compared with the nearest neighbor list of a second item. Based on a result of the comparison, the first and second items can be associated in a cluster.

Abstract: Aspects of the subject matter described herein relate to functions used for retrieving image results based on search queries. More specifically, image search queries can be pre-grouped or classified based on visual and semantic similarity. For example, a pairwise image similarity value for a pair of queries can be computed based on one or more of the sum of all of the overlapping the image results, the sum of the image distances between all of the pairs of images in the image results, and the rank of each of the images in the image results. The pairwise image similarity values can then be used to generate image query clusters. Each image query clusters can include a set of queries with high pairwise image similarity values. In some examples, a distance function can be determined for each image query cluster. This data can be used to provide image results.

Abstract: A method includes receiving search queries. Images are generated that represent the volume of each search query relative to all search queries over a period of time. The images are processed to generate wavelets for each image. Related queries having similar increases or decreases in volume are identified by comparing at least some of the wavelets corresponding to a first query to wavelets corresponding to other queries.

Abstract: An image processing system is provided for encoding videos based on example-based compression. To select the dictionary predictor entries to encode a video, the image processing system reduces the complexity of the video by decomposing the video into smaller pieces. By breaking the video into the simpler pieces, it is easier to locate dictionary predictor entries that are similar to the pieces of the video. The image processing system may decompose the video into one more space-time tubes. For each space-time tube, the image processing system selects dictionary predictor entries to encode the tube.

Abstract: Videos are received and indexed based on fingerprints generated for the videos, using lookup keys which correspond to sub portions of the reference fingerprints. Specifically, a set of lookup keys is selected such that the number of reference fingerprints such that the clumping associated with the associated each of the lookup keys is minimized. Clumping occurs when a number of reference identifiers associated with a lookup key exceeds a maximum value.

Abstract: An exemplar dictionary is built from exemplars of digital content for determining predictor blocks for encoding and decoding digital content. The exemplar dictionary organizes the exemplars as clusters of similar exemplars. Each cluster is mapped to a label. Machine learning techniques are used to generate a prediction model for predicting a label for an exemplar. The exemplar dictionary is used to encode digital content. Clusters of exemplars are obtained by applying a prediction model to a target block of digital content for encoding. A predictor block is selected for encoding the target block based on frequency of occurrence of exemplars in the clusters. The target block is encoded using the predictor block.

Abstract: Methods, and systems, including computer programs encoded on computer-readable storage mediums, including a method for providing a graph that includes entity nodes, label nodes and weighted connecting edges. The method comprises computing an aggregated incoming between-entity edge weight for the entity nodes. When there are positively-weighted incoming between-entity edges into the entity node, the method comprises replacing each of the between-entity edge weights by a pre-normalized between-entity edge weights. The method comprises computing an aggregated from-label weight for the entity node. When there are positively-weighted from-label node edges, the method comprises replacing the corresponding label weights by pre-normalized from-label weights. The method comprises determining influence values for a first, second and third influence factors, where the influence factors have values that sum to one.

Abstract: Systems, methods, devices, and computer program products provide social and interactive applications for detecting repeating content in broadcast media. In some implementations, a method includes: generating a database of audio statistics from content; generating a query from the database of audio statistics; running the query against the database of audio statistics to determine a non-identity match; if a non-identity match exists, identifying the content corresponding to the matched query as repeating content.

Abstract: Embodiments of the present invention recite a method for enhancing the quality of visual prompts in and interactive media response system. In one embodiment, a video coder/decoder (codec) used by a thin device is determined. A visual prompt to be displayed on the thin device is accessed and the display parameters of the visual prompt are modified such that at least one character of the visual prompt is aligned with a blocking artifact generated by the video codec.

Abstract: Methods and systems thereof for monitoring the performance of a streaming media server are described. Server-side measurements are performed with the server operating under a load. Client-side measurements are performed along with the server-side measurements. Data from the server-side measurements and data from the client-side measurements are aligned by time.

Abstract: Large-scale matching of videos is performed by matching a set of probe videos against a set of reference videos to determine if they are visually and/or aurally similar. The visual and audio fingerprints of all probe videos and reference videos are divided into subfingerprints, which are divided into LSH bands. The LSH bands of the probe videos are sorted in one list, and the LSH bands of the reference videos are sorted in another list. Then, the two sorted lists are linearly scanned for matching LSH bands. The matching LSH bands are sorted by probe video ID, and each probe video ID is searched to find matches between probe videos and reference videos. Further, an incremental matching process identifies matches as groups of new probe videos and/or new reference videos are added, without unnecessary repetition of matching old probe videos against old reference videos.

Abstract: An image similar to a target image is selected from among a set of candidate images. A set of image classifiers is first generated and used to create a fingerprint for each candidate image. A hash table is generated for each fingerprint segment, and an identifier for each candidate image is stored in each hash table based on the candidate image fingerprint value for the fingerprint segment associated with the hash table. A fingerprint is created for the target image using the set of classifiers. Segments of the target image fingerprints are compared to segments of the candidate image fingerprints using the hash table, and a candidate image similar to the target image is selected based on this comparison.

Abstract: Methods, systems, and computer program products, including computer programs encoded on a computer readable storage medium, for providing content to a user based on the mode of the user. A method includes: identifying a user for targeting content; evaluating usage information for the user to determine targeting information for a plurality of modes associated with the user; receiving a request to deliver content to the user including an identifier associated with the user and information to determine which mode of the plurality of modes the user is operating in; and providing content to the user based on the mode and associated targeting information.

Abstract: Methods, systems, and apparatus, including computer program products, for ranking images are disclosed. An image search subsystem generates an adjustment factor representative of a quality measure of an image relative to a search query. The quality represents a relevance of the image to the query. The adjustment factor can be computed based on relevance data for the image to the query and image similarity data representing a relative similarity between the image and other images relevant to the query. The relevance data can be based on user actions in response to the image being included in search results for the query. The adjustment factor can be scaled based on whether the relevance data and the image similarity data both indicate that the image is relevant to the search query. A relevance score is computed based on the adjustment factor (e.g., a product of the adjustment factor and relevance score).

Abstract: An exemplar dictionary is built from example image blocks for determining predictor blocks for encoding and decoding images. The exemplar dictionary comprises a hierarchical organization of example image blocks. The hierarchical organization of image blocks is obtained by clustering a set of example image blocks, for example, based on k-means clustering. Performance of clustering is improved by transforming feature vectors representing the image blocks to fewer dimensions. Principal component analysis is used for determining feature vectors with fewer dimensions. The clustering performed at higher levels of the hierarchy uses fewer dimensions of feature vectors compared to lower levels of hierarchy. Performance of clustering is improved by processing only a sample of the image blocks of a cluster. The clustering performed at higher levels of the hierarchy uses lower sampling rates as compared to lower levels of hierarchy.

Abstract: A method includes receiving search queries. Images are generated that represent the volume of each search query relative to all search queries over a period of time. The images are processed to generate wavelets for each image. Related queries having similar increases or decreases in volume are identified by comparing at least some of the wavelets corresponding to a first query to wavelets corresponding to other queries.

Abstract: A method including training a plurality of learning systems, each learning system implementing a learning function and having an input and producing an output, initializing one or more data structures, and evaluating a target sample is described. Also described are methods that include initializing one or more data structures and evaluating a target sample for a best match.