Abstract: Techniques for summarizing media are described. A viewer-interaction analyzer receives a media file containing media, the media file including a plurality of segments. A segment of the media file is scored based on interactions of a set of raters. Viewer metrics on the segment of the media file are measured based on interactions with the segment of the media file by a set of viewers. A set of feature vectors are formed based on the measured viewer interactions, where feature vectors in the set of feature vectors are based on interactions of the set of viewers. A model is trained based on the set of feature vectors and the score assigned to the segment of the media file. The model is applied to segments of the media file to generate an interest rating for segments of the media file. An edited media file is generated based on segments of the media file having interest ratings that meet a criterion.

Abstract: A system, computer-implemented method and non-transitory computer-readable medium for automatically searching for images from events is provided. One or more personal identity tags are provided, wherein the personal identity tags relate to identification information for one or more individuals. Next, at least one event group is identified, wherein the event group is a collection of images associated with an event, the collection of images including one or more images tagged with one or more of the provided personal identity tags. A collection of the images for each of the identified event groups is then received.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training scoring models. One method includes storing data identifying a plurality of positive and a plurality of negative training images for a query. The method further includes selecting a first image from either the positive group of images or the negative group of images, and applying a scoring model to the first image. The method further includes selecting a plurality of candidate images from the other group of images, applying the scoring model to each of the candidate images, and then selecting a second image from the candidate images according to scores for the images. The method further includes determining that the scores for the first image and the second image fail to satisfy a criterion, updating the scoring model, and storing the updated scoring model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, are provided for using audio features to classify audio for information retrieval. In general, one aspect of the subject matter described in this specification can be embodied in methods that include the actions of generating a collection of auditory images, each auditory image being generated from respective audio files according to an auditory model; extracting sparse features from each auditory image in the collection to generate a sparse feature vector representing the corresponding audio file; and ranking the audio files in response to a query including one or more words using the sparse feature vectors and a matching function relating sparse feature vectors to words in the query.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training scoring models. One method includes storing data identifying a plurality of positive and a plurality of negative training images for a query. The method further includes selecting a first image from either the positive group of images or the negative group of images, and applying a scoring model to the first image. The method further includes selecting a plurality of candidate images from the other group of images, applying the scoring model to each of the candidate images, and then selecting a second image from the candidate images according to scores for the images. The method further includes determining that the scores for the first image and the second image fail to satisfy a criterion, updating the scoring model, and storing the updated scoring model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying similar images. In some implementations, a method is provided that includes receiving a collection of images and data associated with each image in the collection of images; generating a sparse feature representation for each image in the collection of images; and training an image similarity function using image triplets sampled from the collection of images and corresponding sparse feature representations.

Abstract: This disclosure is directed to providing audio playback to a mobile device user. According to one aspect of this disclosure, a mobile device may be to modify audio playback in response to detecting an inclination of the mobile device (and thereby a user) with respect to a reference plane. According to another aspect of this disclosure, a mobile device may be configured to automatically identify an audible sound that may be motivational to a user, and store an indication of the audible sound in response to the identification. According to another aspect of this disclosure, a mobile device may automatically play back a previously identified motivational song in response to detection of user movement.

Abstract: A method of decomposing a set of scans of different views of overlapping objects into constituent objects is presented. The method involves an initialization process whereby keypoints in two views are determined and matched, disparity between keypoint pairs are computed, and the keypoints are grouped into clusters based on their disparities. Following the initialization process is an iterative optimization process whereby a cost function is calculated and minimized assuming a fixed composition matrix and re-solved assuming a fixed attenuation coefficient. Then, the composition matrix and the attenuation coefficient are updated simultaneously, and the solving, the re-solving, and the updating steps are repeated until there is no significant improvement in the result.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training scoring models. One method includes storing data identifying a plurality of positive and a plurality of negative training images for a query. The method further includes selecting a first image from either the positive group of images or the negative group of images, and applying a scoring model to the first image. The method further includes selecting a plurality of candidate images from the other group of images, applying the scoring model to each of the candidate images, and then selecting a second image from the candidate images according to scores for the images. The method further includes determining that the scores for the first image and the second image fail to satisfy a criterion, updating the scoring model, and storing the updated scoring model.

Abstract: A system, computer readable storage medium, and computer-implemented method presents video search results responsive to a user keyword query. The video hosting system uses a machine learning process to learn a feature-keyword model associating features of media content from a labeled training dataset with keywords descriptive of their content. The system uses the learned model to provide video search results relevant to a keyword query based on features found in the videos. Furthermore, the system determines and presents one or more thumbnail images representative of the video using the learned model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, are provided for using audio features to classify audio for information retrieval. In general, one aspect of the subject matter described in this specification can be embodied in methods that include the actions of generating a collection of auditory images, each auditory image being generated from respective audio files according to an auditory model; extracting sparse features from each auditory image in the collection to generate a sparse feature vector representing the corresponding audio file; and ranking the audio files in response to a query including one or more words using the sparse feature vectors and a matching function relating sparse feature vectors to words in the query.

Abstract: A method of decomposing a set of scans of different views of overlapping objects into constituent objects is presented. The method involves an initialization process whereby keypoints in two views are determined and matched, disparity between keypoint pairs are computed, and the keypoints are grouped into clusters based on their disparities. Following the initialization process is an iterative optimization process whereby a cost function is calculated and minimized assuming a fixed composition matrix and re-solved assuming a fixed attenuation coefficient. Then, the composition matrix and the attenuation coefficient are updated simultaneously, and the solving, the re-solving, and the updating steps are repeated until there is no significant improvement in the result.

Abstract: A method and system for scheduling events into a set of opportunities is presented. The method includes 1) dividing a path of an image acquisition device so that there is at least a first portion and a second portion at any given moment, wherein each of the first portion and the second portion includes at least one state and the first portion includes a null state in which no image is taken; 2) combining each state in the first portion with at least one state in the second portion one by one to generate a series of updated sequences; and 3) selecting at least one of the updated sequences based on a merit value associated with each of the updated sequences. The invention uses only two groups out of all the relevant opportunities for most calculations, and is especially applicable to situations like satellite pass scheduling.

Abstract: A method and system for scheduling events into a set of opportunities is presented. The method includes 1) dividing a path of an image acquisition device so that there is at least a first portion and a second portion at any given moment, wherein each of the first portion and the second portion includes at least one state and the first portion includes a null state in which no image is taken; 2) combining each state in the first portion with at least one state in the second portion one by one to generate a series of updated sequences; and 3) selecting at least one of the updated sequences based on a merit value associated with each of the updated sequences. The invention uses only two groups out of all the relevant opportunities for most calculations, and is especially applicable to situations like satellite pass scheduling.