Method and system for concept summarization

A computer program product stored on a non-transitory computer-readable medium that when executed by a processor, performs a method for summarizing a concept, the method comprising:

• receiving, from a user, a query corresponding to the concept;
• collecting a plurality of images along with descriptive information corresponding thereto, the plurality of images being collected from a source corresponding to the query;
• processing the plurality of images and the descriptive information to form feature vectors and processed descriptive information respectively, each of the plurality of images corresponding to one of the feature vectors;
• assigning each of the plurality of images with one or more topic distribution values corresponding to one or more topics, the one or more topics corresponding to the processed descriptive information;
• receiving, from the user, an input that includes a summary size that represents a number of images for the summarizing;
• determining a sparse set of images from the plurality of images to summarize the concept, the sparse set of images being determined based on the feature vectors and the assigned topic distribution values, wherein a size of the sparse set of images is limited to the summary size;
• regularizing one or more distribution constraints for building a target summary from the summarized concept by receiving constraints from the user that are used by the summarizing module when determining the sparse set of images, wherein the input received from the user includes the constraints;
• and
• providing the sparse set of images responsive to the query.

A system for summarizing a concept, the system comprising a processor and a memory for:

• receiving a query from a user, wherein the query is corresponding to the concept;
• collecting a plurality of images and descriptive information corresponding to each of the plurality of images from a source, the source being identified based on the query, wherein the plurality of images correspond to the query;
• converting the plurality of images into corresponding feature vectors, processing the descriptive information corresponding to each image of the plurality of images;
• identifying one or more topics for the plurality of images based on the processed descriptive information;
• assigning one or more topic distribution values corresponding to the one or more topics to each image of the plurality of images;
• determining a sparse set of images from the plurality of images to summarize the concept, the sparse set of images being determined based on the corresponding feature vectors and the assigned topic distribution values, wherein the summary of the concept produced by the summarizing module is related to a summary size representing a number of images in the sparse set of images, and further wherein the summary size is input received from the user;
• regularizing one or more distribution constraints for the summarizing module to build a target summary from the summarized concept, the one or more distribution constraints being regularized by utilizing a regularization term to minimize divergence between the summarized concept and the target summary, wherein the regularization term is received as part of the input from the user;
• and
• providing the sparse set of images responsive to the query.