758results about "Vectoral format still image data" patented technology

In the past, databases containing three dimensional images were searched by keywords and indexes. For such functionality, a set of keywords must be associated with each image. This is usually performed by a person and is prone to errors. It has now been found that searching of three-dimensional images stored electronically can be performed in dependence upon a variety of criteria such as colour, shape, size, etc. A method is disclosed for searching three dimensional images of objects in dependence upon the above criteria, other images, or other criteria. This obviates the need for keyword indexing of image databases.

A method and system for visually annotating an image. Annotations and notes to images, such as digital medical and healthcare images, may be stored in a structured vector representation alongside image information in a single, non-volatile and portable file or in a separate file from the image. The annotations may be composed of point, line and polygon drawings and text symbols, labels or definitions and captions or descriptions. The annotations may be structured in a manner that facilitates grouping and manipulation as user defined groups. The annotations may be related to an image but not inextricably bound such that the original image is completely preserved. Annotations may further be selectively displayed on the image for context appropriate viewing. The annotations may be retrieved for purposes such as editing, printing, display, indexing and reporting for example, and may be displayed on an image for interactive use with an embedded self-contained user interface.

System and method for performing Collaborative Filtering while preserving complete user anonymity are provided. Each of a group of client devices sends a rating vector anonymously to a server. The cells in each rating vector correspond to a set of items, and selected cells have ratings provided by the user associated with the corresponding client device for the corresponding items. The server aggregates all the rating vectors into a rating matrix, and factorizes the rating matrix into a user feature matrix and an item feature matrix through approximation, such that the rating matrix equals the product of the user feature matrix and the item feature matrix. The item feature matrix is sent to the client devices. Each of the client devices calculates its own user feature vector based on its rating vector and the item feature matrix, and provides personalized recommendations on selected items based on the client's user feature vector and the item feature matrix.

The invention provides an information recommendation method. The method comprises the steps of obtaining image information, and according to the image information, generating a user tag set corresponding to user identifiers, wherein the image information has the corresponding user identifiers; according to user tags in the user tag set and the corresponding image information, forming eigenvectorscorresponding to the user identifiers; inputting the eigenvectors to a trained information recommendation model, and outputting recommendation parameters of users for to-be-recommended information; and according to the recommendation parameters, recommending the to-be-recommended information to terminals corresponding to the user identifiers. The user tag set capable of reflecting user features isestablished through the image information; the recommendation parameters of the users for the recommended information are predicted by adopting the trained information recommendation model accordingto the user tag set; and the recommendation is performed according to the recommendation parameters, so that the recommendation accuracy is improved and the users not interested in the information areprevented from being disturbed. In addition, the invention provides an information recommendation apparatus, a computer device and a storage medium.