A detection device for detecting human-body orientation includes a camera and a processing device. The camera is configured to capture a human-body image. The processing device is configured to cut a human head contour image in the human-body image to obtain an input image, and input the input image to a classifier. The classifier outputs a plurality of human-body orientation probabilities for the input image. The processing device finds the highest human-body orientation probability, and determines whether the highest human-body orientation probability is above the accuracy threshold. In response to the highest human-body orientation probability being above the accuracy threshold, the processing device regards the human-body orientation corresponding to the highest human-body orientation probability as the determined human-body orientation.

A first period when detection of a detection target body in contact with or proximity to a sensor is performed and a second period when detection of asperities on a surface of the detection target body is performed are set. First regions are set in a detection region of the sensor as input keys for a password. In the first period, one of segmented regions obtained by dividing a second region the center of which corresponds to coordinates of a touch detection position when the touch detection position for inputting of the password is detected in one of the first regions. In the second period, based on priorities set in advance according to the number of feature points of the detection target body, a segmented image detected in the segmented region for the inputting is output, and a certain number of segmented images are synthesized to generate a fingerprint image.

Disclosed are a method and apparatus for identifying versions of a form. In an example, clients of a medical company fill out many forms, and many of these forms have multiple versions. The medical company operates in 10 states, and each state has a different version of a client intake form, as well as of an insurance identification form. In order to automatically extract information from a particular filled out form, it may be helpful to identify a particular form template, as well as the version of the form template, of which the filled out form is an instance. A computer system evaluates images of filled out forms, and identifies various form templates and versions of form templates based on the images.

A method, apparatus and system for zero shot object detection includes, in a semantic embedding space having embedded object class labels, training the space by embedding extracted features of bounding boxes and object class labels of labeled bounding boxes of known object classes into the space, determining regions in an image having unknown object classes on which to perform object detection as proposed bounding boxes, extracting features of the proposed bounding boxes, projecting the extracted features of the proposed bounding boxes into the space, computing a similarity measure between the projected features of the proposed bounding boxes and the embedded, extracted features of the bounding boxes of the known object classes in the space, and predicting an object class label for proposed bounding boxes by determining a nearest embedded object class label to the projected features of the proposed bounding boxes in the space based on the similarity measures.

An image processor includes an imaging device that captures an image of a road surface around a vehicle V, and a control portion that detects a marker drawn on the road surface from the captured image. The control portion connects a plurality of broken markers to create a single marker when the detected marker is broken into plural.

Technologies are described for retrieving documents using image representations in the documents and is based on intra-image features. The identification of elements within an image representation can allow for deeper understanding of the image representation and for better relating image representations based on their intra-image features. The intra-image features present in image representations can be used in searches. Search results can further be reranked to improve search results. For example, reranking can allow search results to conform to intra-image dominant image features.

A computer-implemented method of estimating a 6D pose and shape of one or more objects from a 2D image, comprises the steps of: detecting, within the 2D image, one or more 2D regions of interest, each 2D region of interest containing a corresponding object among the one of more objects; cropping out a corresponding pixel value array, coordinate tensor , and feature map for each 2D region of interest; concatenating the corresponding pixel value array, coordinate tensor, and feature map for each 2D region of interest; and inferring, for each 2D region of interest, a 4D quaternion describing a rotation of the corresponding object in the 3D rotation group, a 2D centroid, which is a projection of a 3D translation of the corresponding object onto a plane of the 2D image given a camera matrix associated to the 2D, image, a distance from a viewpoint of the 2D image to the corresponding object a size and a class-specific latent shape vector of the corresponding object.

A computer acquires training data, in which first image data, object information indicating first objects included in the first image data, and relationship information indicating a first relationship between the first objects are associated. The computer executes machine learning that trains, based on the training data, an inference model that infers both second objects included in second image data and a second relationship between the second objects or selectively infers one of the second objects and the second relationship according to an input of the second image data to the inference model. The machine learning uses a penalty term when calculating an error between an inference result of the inference model and the training data. The penalty term causes the error to increase as an overlap between inferred image regions, which are inferred to be image regions in which objects are present in the inference result, increases.

Techniques are disclosed for characterizing and defining the location of a copy space in an image. A methodology implementing the techniques according to an embodiment includes applying a regression convolutional neural network (CNN) to an image. The regression CNN is configured to predict properties of the copy space such as size and type (natural or manufactured). The prediction is conditioned on a determination of the presence of the copy space in the image. The method further includes applying a segmentation CNN to the image. The segmentation CNN is configured to generate one or more pixel-level masks to define the location of copy spaces in the image, whether natural or manufactured, or to define the location of a background region of the image. The segmentation CNN may include a first stage comprising convolutional layers and a second stage comprising pairs of boundary refinement layers and bilinear up-sampling layers.

A method of price text detection by an imaging controller comprises obtaining, by the imaging controller, an image of a shelf supporting labels bearing price text, generating, by the imaging controller, a plurality of text regions containing candidate text elements from the image, assigning, by the imaging controller, a classification to each of the text regions, selected from a price text classification and a non-price text classification. The imaging controller, within each of a subset of the text regions having the price text classification: detects a price text sub-region and generates a price text string by applying character recognition to the price text sub-region. The method further includes presenting, by the imaging controller, the locations of the subset of text regions, in association with the corresponding price text strings.