A candidate region extraction unit (110) extracts candidate regions from an image by using a common characteristic amount. A part region extraction unit (120) separates a part region, which has a high likelihood of being a part, from a second-part candidate region other than the part region. A characteristic amount calculation unit (130) calculates the intrinsic characteristic amount of each of the part region and the second-part candidate region. A characteristic amount correction unit (140) corrects the intrinsic characteristic amount of either one of the candidate regions on the basis of the length, the width, and the angle of the other candidate region. A determination unit (150) determines whether or not the part region and the second-part candidate region are parts that constitute a pair on the basis of the similarity between the corrected intrinsic characteristic amount and the intrinsic characteristic amount of the other candidate region.

A method for visual recognition of an object in an electronic image includes extracting unique points of an object to be learned and/or a target object. The unique points are obtained by cross-correlating the image with a structure. Generally, the structure and/or the size of the structure may vary to detect extremum information associated with the learned object and/or target object. An icon corresponding to each of the unique points is extracted. The size of the icon corresponds to the scale of the unique point. After extraction of the various icons, an object becomes a collection of icons. Each of these icons is un-rotated and normalized or resized to a constant size so it can be compared with other icons. One of the unique properties of these icons is their stability over scale and angle. Thus, this invention allows the recognition of an image(s) or object(s) from large number of trained images or objects very quickly.

Systems and methods provide adapted content to a visitor to a physical environment. An example method receives an image of a visitor to an environment. A visitor portion of the image is distinct from an environment portion of the image. The method detects one or more shapes in the visitor portion of the image using an automatic shape detection technique and defines an approximate boundary of the one or more shapes using a mask. The one or more shapes can be shapes of the visitor's clothing items. The method then calculates an attribute for an area of the image within the mask and identifies electronic content based on the attribute for the area of the image within the mask. The attribute can be a color attribute for the area such as a median color or a dominant color. The method provides the identified electronic content for display in the environment.

A method semantically labels an image acquired of a scene by first obtaining a local semantic feature for each local region in the image. The local semantic features are combined recursively to form intermediate segments until a semantic feature for the entire image is obtained. Then, the semantic feature for the entire image is decombined recursively into intermediate segments until an enhanced semantic feature for each local region is obtained. Then, each local region is labeled according to the enhanced semantic feature.

Multi-source multi-modal activity recognition for conducting aerial video surveillance comprising detecting and tracking multiple dynamic targets from a moving platform, representing FMV target tracks and chat-messages as graphs of attributes, associating FMV tracks and chat-messages using a probabilistic graph based mapping approach; and detecting spatial-temporal activity boundaries.

A method includes receiving a first image that is captured at a first time. The method also includes detecting a location of a first object in the first image. The method also includes determining a region of interest based at least partially upon the location of the first object in the first image. The method also includes receiving a second image that is captured at a second time. The method also includes identifying the region of interest in the second image. The method also includes detecting a location of a second object in a portion of the second image that is outside of the region of interest.

This invention focuses specifically on the use of epipolar lines and the use of matrix transformations to coordinate cameras. This invention organizes cameras in a manner which is intuitive and effective in perceiving perspectives which are not normally possible; to calculate range precisely; to allow redundancy; to corroborate feature recognition; and to allow perspectives from angles from which no cameras exist. By enabling remote scene reconstruction with a limited set of images, transmission bandwidth is greatly conserved.

The object detection means, when given a frame, derives a bounding box of a detection target object from within the frame based on a learning model that includes multiple layers, and defines output data of one layer in the learning model or the frame itself as a feature value of the frame. The feature value acquisition means acquires a feature value of the bounding box of an attribute identification target object from the feature value of the frame. The similarity degree determination means determines a similarity degree, and when the similarity degree is equal to or greater than a predetermined threshold, stops extracting the one or more attributes of the attribute identification target object, and identifies the one or more attributes of the attribute identification target object in the bounding box by diverting the one or more attributes corresponding to the feature value of the past bounding box.

A system and method of automatic product attribute recognition receive training images having bounding boxes associated with one or more products in the training images, receive attribute values for each of the one or more products in the training images, and train a first convolutional neural network (CNN) model to generate bounding boxes for and identify each of the one or more products with the training images until the accuracy of the first CNN model is above a first predetermined threshold. The system and method further train a second CNN model for each of the products associated with the cropped images until the second CNN generates attribute values for the one or more attributes with an accuracy above a second predetermined threshold, and automatically recognize the one or more attributes for a new product image by presenting the product image to the first and second CNN models.

A system and method of automatic product attribute recognition receive training images having bounding boxes associated with one or more products in the training images, receive attribute values for each of the one or more products in the training images, and train a first convolutional neural network (CNN) model to generate bounding boxes for and identify each of the one or more products with the training images until the accuracy of the first CNN model is above a first predetermined threshold. The system and method further train a second CNN model for each of the products associated with the cropped images until the second CNN generates attribute values for the one or more attributes with an accuracy above a second predetermined threshold, and automatically recognize the one or more attributes for a new product image by presenting the product image to the first and second CNN models.