Layout support system, layout support apparatus, information accumulation apparatus, and storage medium
AssigneeNEC Corporation
InventorTakahiro Toizumi
Life of this patent
Estimating the term…
Abstract
This invention contributes to improvement in efficiency in generation of a layout. A layout support system includes: an information accumulation apparatus that extracts image feature vectors of a plurality of existing contents and stores, in a database, the image feature vectors in association with pieces of layout information; and a layout support apparatus that detects, on the basis of similarities between the image feature vectors stored in the database and a comparative feature vector generated for a target content, a piece of layout information matching the target content, from among the pieces of layout information stored in the database. The layout support system can support decision-making regarding layout determination.
Claims8
A layout support system comprising:
- an information accumulation apparatus that includes at least one processor, the at least one processor executing, for each of a plurality of existing contents, a process of extracting an image feature vector from an image indicating a layout of constituent elements of the existing content and orders of priority set to the constituent elements and a process of storing, in a database, the image feature vector and a piece of layout information indicating the layout of the constituent elements in such a manner that the image feature vector and the piece of layout information are in association with each other;
- and
- a layout support apparatus that includes at least one processor, the at least one processor executing a process of calculating similarities between the image feature vectors stored in the database and a comparative feature vector generated on a basis of target content information indicating at least orders of priority set to constituent elements constituting a target content which is a target to be laid out and a process of detecting, on a basis of the similarities, a piece of layout information matching the target content, from among the pieces of layout information stored in the database,
- wherein the at least one processor of the layout support apparatus is further configured to:
- extract, from the target content information, feature vectors respectively corresponding to the constituent elements of the target content;
- mix the feature vectors to generate a plurality of mixed feature vectors;
- integrate the plurality of mixed feature vectors to generate an integrated feature vector of a given data size;
- and
- generate the comparative feature vector on a basis of the integrated feature vector.
A layout support apparatus comprising at least one processor, the at least one processor executing:
- for each of a plurality of existing contents, a feature vector comparing process of
- (a) referring to a database in which pieces of layout information each indicating a layout of constituent elements constituting a corresponding one of the plurality of existing contents and image feature vectors each extracted from an image indicating the layout and orders of priority set to the constituent elements are stored in association with each other;
- and
- (b) calculating similarities between the image feature vectors stored in the database and a comparative feature vector generated on a basis of target content information indicating at least orders of priority set to constituent elements constituting a target content which is a target to be laid out;
- a detecting process of detecting, on a basis of the similarities, a piece of layout information matching the target content, from among the pieces of layout information of the plurality of existing contents;
- and
- a comparative feature vector generating process comprising:
- extracting, from the target content information, feature vectors respectively corresponding to the constituent elements of the target content;
- mixing the feature vectors to generate a plurality of mixed feature vectors;
- integrating the plurality of mixed feature vectors to generate an integrated feature vector of a given data size;
- and
- generating the comparative feature vector on a basis of the integrated feature vector.
Dependencies are inferred from the claim text.