A translation system provides machine translations of review texts on item pages using context from the item pages outside of the review text being translated. Given review text from an item page, context for machine translating the review text is determined from the item page. In some aspects, one or more keywords are determined based on text, images, and/or videos on the item page. The one or more keywords are used as context by the machine translator to translate the review text from a first language to a second language to provide translated review text, which can be presented on the item page.

An image processing system has one or more memories and image processing circuitry coupled to the one or more memories. The image processing circuitry, in operation, compares a first image to feature data in a comparison image space using a matching model. The comparing includes: unwarping keypoints in keypoint data of the first image; and comparing the unwarped keypoints and descriptor data associated with the first image to the feature data of the comparison image. The image processing circuitry determines whether the first image matches the comparison image based on the comparing.

An apparatus obtains first transformation information, such as a first transformation matrix, to be used for coordinate transformation between a coordinate system in an overall image prepared beforehand and a coordinate system in a first captured image, by comparing a feature point extracted from the overall image and a feature point extracted from the first captured image. In a case where the first transformation information is updated, the apparatus generates a partial image from the overall image based on an image-taking position of a just preceding image, and compares a feature point extracted from the partial image with a feature point extracted from a captured image to be used for updating of the first transformation information, and accordingly obtains transformation information for updating. The apparatus updates the first transformation information by using the obtained transformation information for updating. Thus, accuracy of tracking processing is improved.

Apparatus and methods to unwarp at least portions of distorted, electronically-captured images are described. Keypoints, instead of an entire image, may be unwarped and used in various machine-vision algorithms, such as object recognition, image matching, and 3D reconstruction algorithms. When using unwarped keypoints, the machine-vision algorithms may perform reliably irrespective of distortions that may be introduced by one or more image capture systems.

A method of customized installation of access control hardware according to one embodiment includes capturing, by a camera of a mobile device, at least one image of an installation location for the access control hardware, generating a set of customized installation instructions for the access control hardware at the installation location based on the at least one image, and displaying the customized installation instructions on a graphical user interface of the mobile device.

During an image-analysis technique, the system calculates features by performing image analysis (such as optical character recognition) on a received image of a document. Using these features, as well as spatial and grammatical constraints, the system determines a layout of the document. For example, the layout may be determined using constraint-based optimization based on the spatial and the grammatical constraints. Note that the layout specifies locations of content in the document, and may be used to subsequently extract the content from the image and/or to allow a user to provide feedback on the extracted content by presenting the extracted content to the user in a context (i.e., the determined layout) that is familiar to the user.

A location providing method implemented by a server comprising extracting text from an image received through an application executed on a first client, extracting matched location information corresponding to the image based on the text, and providing the matched location information and the image to one or more clients through the application.

A method of detecting fields in document images includes: receiving a codebook comprising a set of visual words, each visual word corresponding to a center of a cluster of local descriptors; calculating, based on a set of user labeled document images, for each visual word of the codebook, a respective frequency distribution of a field position of a specified labeled field with respect to the visual word; loading a document image for extraction of target fields; calculating a statistical predicate of a possible position of a target field in the document image based on the frequency distributions; and detecting, using the trained model, fields in the document image based on the calculated statistical predicate.

A method of generating and optimizing a codebooks for document analysis comprises: receiving a first set of document images; extracting a plurality of keypoint regions from each document image of the first set of document images; calculating local descriptors for each keypoint region of the extracted keypoint regions; clustering the local descriptors such that each center of a cluster of local descriptors corresponds to a respective visual word; generating a codebook containing a set of visual words; and optimizing the codebook by maximizing mutual information (MI) between a target field of a second set of document images and at least one visual word of the set of visual words.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for an active lighting system. In one aspect, a method includes receiving a first image of the physical document having a first glare signature and a second image of the physical document having a second glare signature that is different from the first glare signature; determining a first glare map of the first image and a second glare map of the second image; comparing the first glare map to the second glare map; and generating the digital image based on the comparison of the first and second glare maps.