(57) ABSTRACT A method of modelling data, comprising: training an objec tive function of a
linear classifier, based on a set of labeled data, to derive a set of classifier weights; defining …

In techniques for image foreground detection, a foreground detection module is
implemented to generate varying levels of saliency thresholds from a saliency map of an …

In techniques for iterative saliency map estimation, a salient regions module applies a
saliency estimation technique to compute a saliency map of an image that includes image …

An image application may enable users to capture images on cameras of their mobile
devices, which the image application can then submit as queries to perform a search …

The trend of analyzing big data in artificial intelligence requires more scalable machine
learning algorithms, among which clustering is a fundamental and arguably the most widely …

US9158988B2 - Method for detecting a plurality of instances of an object - Google Patents
US9158988B2 - Method for detecting a plurality of instances of an object - Google Patents …

BACKGROUND The use in commerce of network-based systems such as the Internet has
grown exponentially over the last decade or so. Such network-based based systems can be …

Computationally efficient accelerated D2-clustering algorithms are disclosed for clustering
discrete distributions under the Wasserstein distance with improved scalability. Three first …

An image processing system acquires a digital image and determines notional weights for
each pixel in the image. The notional weight of each pixel relates to the pixel's color value of …

In techniques of combined composition and change-based models for image cropping, a
composition application is implemented to apply one or more image composition modules of …