Abstract: In an embodiment, the present invention relates to a method for semantic analysis of digital multimedia. In an embodiment of the invention, low level features are extracted representative of one or more concepts. A discriminative classifier is trained using these low level features. A collective annotation model is built based on the discriminative classifiers. In various embodiments of the invention, the frame work is totally generic and can be applied with any number of low-level features or discriminative classifiers. Further, the analysis makes no domain specific assumptions, and can be applied to activity analysis or other scenarios without modification. The framework admits the inclusion of a broad class of potential functions, hence enabling multi-modal analysis and the fusion of heterogeneous information sources.

Abstract: Embodiments of the present invention introduce a user navigation interface that allows a user to monitor/navigate video streams captured from multiple cameras. It integrates video streams from multiple cameras with the semantic layout into a 3-D immersive environment and renders the video streams in multiple displays on a user navigation interface. It conveys the spatial distribution of the cameras as well as their fields of view and allows a user to navigate freely or switch among preset views. This description is not intended to be a complete description of, or limit the scope of, the invention. Other features, aspects, and objects of the invention can be obtained from a review of the specification, the figures, and the claims.

Abstract: Embodiments of the present invention introduce a novel technique to analyze and monitor video streams captured from multiple cameras. It highlights the foreground region of the video streams via local alpha blending and displays the videos in an immersive 3-D environment. The spatial arrangement of the displays can be generated by multi-dimensional scaling of the amount of simultaneous motion across different video streams. This description is not intended to be a complete description of, or limit the scope of, the invention. Other features, aspects, and objects of the invention can be obtained from a review of the specification, the figures, and the claims.

Abstract: Techniques for media segmentation are disclosed. A number of measures used to generate similarity values are compared. Two different approaches to calculate feature vectors based on pairwise frame similarity in combination with different kernel functions are disclosed. A supervised classification method is used with the feature vectors to assess segment boundaries.

Abstract: Embodiments of the present invention provide a system and method for discriminatively selecting keyframes that are representative of segments of a source digital media and at the same time distinguishable from other keyframes representing other segments of the digital media. The method and system, in one embodiment, includes pre-processing the source digital media to obtain feature vectors for frames of the media. Discriminatively selecting a keyframe as a representative for each segment of a source digital media wherein said discriminative selection includes determining a similarity measure for each candidate keyframe and determining a dis-similarity measure for each candidate keyframe and selecting the keyframe with the highest goodness value computing from the similarity and dis-similarity measures.

Abstract: Techniques for reducing the computational complexity of conventional similarity-based approaches for temporal event clustering of digital photograph collections include one or more approaches to select boundaries based on dynamic programming and the Bayes information criterion. Each method performs competitively with conventional approaches and offer significant computational savings.

Abstract: Described is a system for automatic digital photo orientation detection. We leverage online public photos with great content variation to extract effective features with layout information. Classification proceeds using an approximate nearest neighbors approach which scales well to massive training sets, hardly compromising efficiency. We have tested the method successfully on the largest data set to date of nearly 30,000 Flickr photos as well as both difficult and typical consumer usage scenarios. Though limited data are available for comparison across different systems, the proposed system significantly outperforms a state of the art system on a common data set.

Abstract: A stream of ordered information, such as, for example, audio, video and/or text data, can be windowed and parameterized. A similarity between the parameterized and windowed stream of ordered information can be determined, and a probabilistic decomposition or probabilistic matrix factorization, such as non-negative matrix factorization, can be applied to the similarity matrix. The component matrices resulting from the decomposition indicate major components or segments of the ordered information. Excerpts can then be extracted from the stream of ordered information based on the component matrices to generate a summary of the stream of ordered information.

Abstract: The present invention provides a system and method for automatically combining image and audio data to create a multimedia presentation. In one embodiment, audio and image data are received by the system. The audio data includes a list of events that correspond to points of interest in an audio file. The audio data may also include an audio file or audio stream. The received images are then matched to the audio file or stream using the time. In one embodiment, the events represent times within the audio file or stream at which there is a certain feature or characteristic in the audio file. The audio events list may be processed to remove, sort or predict or otherwise generate audio events. Images processing may also occur, and may include image analysis to determine image matching to the event list, deleting images, and processing images to incorporate effects. Image effects may include cropping, panning, zooming and other visual effects.

Abstract: In an embodiment, the present invention relates to a method for semantic analysis of digital multimedia. In an embodiment of the invention, low level features are extracted representative of one or more concepts. A discriminative classifier is trained using these low level features. A collective annotation model is built based on the discriminative classifiers. In various embodiments of the invention, the frame work is totally generic and can be applied with any number of low-level features or discriminative classifiers. Further, the analysis makes no domain specific assumptions, and can be applied to activity analysis or other scenarios without modification. The framework admits the inclusion of a broad class of potential functions, hence enabling multi-modal analysis and the fusion of heterogeneous information sources.

Abstract: Methods and systems for classifying images, such as photographs, allow a user to incorporate subjective judgments regarding photograph qualities when making classification decisions. A slide-show interface allows a user to classify and advance photographs with a one-key action or a single interaction event. The interface presents related information relevant to a displayed photograph that is to be classified, such as contiguous photographs, similar photographs, and other versions of the same photograph. The methods and systems provide an overview interface which allows a user to review and refine classification decisions in the context of the original sequence of photographs.

Abstract: Embodiments of the present invention provide a method for producing a summary of a digital file on one or more computers. The method includes segmenting the digital file into a plurality of segments, clustering said segments into a plurality of clusters and selecting a cluster from said plurality of clusters wherein said selected cluster includes segments representative of said digital file. Upon selection of a cluster a segment of the cluster is provided as a summary of said digital file.

Abstract: A method of extracting audio excerpts comprises: segmenting audio data into a plurality of audio data segments; setting a fitness criteria for the plurality of audio data segments; analyzing the plurality of audio data segments based on the fitness criteria; and selecting one of the plurality of audio data segments that satisfies the fitness criteria. In various exemplary embodiments, the method of extracting audio excerpts further comprises associating the selected one of the plurality of audio data segments with video data. In such embodiments, associating the selected one of the plurality of audio data segments with video data may comprise associating the selected one of the plurality of audio data segments with a keyframe.

Abstract: Methods and systems for classifying images, such as photographs, allow a user to incorporate subjective judgments regarding photograph qualities when making classification decisions. A slide-show interface allows a user to classify and advance photographs with a one-key action or a single interaction event. The interface presents related information relevant to a displayed photograph that is to be classified, such as contiguous photographs, similar photographs, and other versions of the same photograph. The methods and systems provide an overview interface which allows a user to review and refine classification decisions in the context of the original sequence of photographs.

Abstract: Embodiments of the present invention provides a method for producing a summary of a digital file. The method includes segmenting the digital file into a plurality of segments, clustering said segments into a plurality of clusters and selecting a cluster from said plurality of clusters wherein said selected cluster includes segments representative of said digital file. Upon selection of a cluster a segment of the cluster is provided as a summary of said digital file.

Abstract: The present invention provides a system and method for automatically combining image and audio data to create a multimedia presentation. In one embodiment, audio and image data are received by the system. The audio data includes a list of events that correspond to points of interest in an audio file. The audio data may also include an audio file or audio stream. The received images are then matched to the audio file or stream using the time. In one embodiment, the events represent times within the audio file or stream at which there is a certain feature or characteristic in the audio file. The audio events list may be processed to remove, sort or predict or otherwise generate audio events. Images processing may also occur, and may include image analysis to determine image matching to the event list, deleting images, and processing images to incorporate effects. Image effects may include cropping, panning, zooming and other visual effects.

Abstract: A method for measuring the similarity between the beat spectra of two or more audio works. A distance formula is used to measure the similarity by rhythm and tempo between shortened beat spectra B1(L) and B2(L). The result is a vector which measures the similarity of rhythm and tempo. A distance formula is used to measure the rhythmic similarity between the scaled beat spectra B1(L) and B2(L). The result is a measure of rhythmically similar music regardless of the tempo. The method can be used in a wide variety of applications, including concatenating music with similar tempos, automatic music sequencing, classification of music into genres, search for music with similar rhythmic structures, search for music with similar rhythmic and tempo structures, and ranking music according to a similarity measure.

Abstract: A method of extracting audio excerpts comprises: segmenting audio data into a plurality of audio data segments; setting a fitness criteria for the plurality of audio data segments; analyzing the plurality of audio data segments based on the fitness criteria; and selecting one of the plurality of audio data segments that satisfies the fitness criteria. In various exemplary embodiments, the method of extracting audio excerpts further comprises associating the selected one of the plurality of audio data segments with video data. In such embodiments, associating the selected one of the plurality of audio data segments with video data may comprise associating the selected one of the plurality of audio data segments with a keyframe.