Generating and ordering tags for an image using subgraph of concepts
AssigneeInternational Business Machines Corporation
InventorsBishwaranjan Bhattacharjee, Tushar Nagarajan
Life of this patent
Estimating the term…
Abstract
Aspects include a system, computer program production and computer-implemented method for tagging an image. An image classification engine stored in a memory of a computer device generates a plurality of tags for the image and uses the plurality of tags to generate a relevance subgraph for the image. An embedding engine embeds nodes and edges of the relevance subgraph into fixed dimension vectors of a matrix. A neural network stored in the memory determines a feature vector from the image. A processor applies the feature vector to the matrix to generate a context vector for the image. The context vector is used to tag the image.
Claims20
A computer-implemented method for tagging an image, the computer-implemented method comprising:
- generating a relevance subgraph for the image using tags for the image generated by an image classification engine stored in one or more memories of the one or more computing devices, the image classification engine being used to generate a plurality of tags for the image;
- embedding, using an embedding engine, the nodes and edges of the relevance subgraph into fixed dimension vectors of a matrix;
- determining, using a neural network stored in one or more memories of the one or more computing devices, the neural network engine used to generate, a feature vector from the image;
- applying, using a processor, the feature vector to the matrix to generate a context vector for the image;
- and
- using the context vector to tag the image.
A system for tagging an image, the system comprising:
- a memory having computer readable instructions;
- and
- one or more processors for executing the computer readable instructions, the computer readable instructions configured to perform a method comprising:
- generating a relevance subgraph from the image, the subgraph having a plurality of nodes and edges;
- embedding the nodes and edges of the relevance subgraph in a fixed dimension vector;
- determining a feature vector from the image;
- applying the feature vector to the fixed dimension vector to generate a context vector for the image;
- and
- using the context vector to tag the image.
A computer program product for tagging an image, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform:
- generating a relevance subgraph from the image, the subgraph having a plurality of nodes and edges;
- embedding the nodes and edges of the relevance subgraph in a fixed dimension vector;
- determining a feature vector from the image;
- applying the feature vector to the fixed dimension vector to generate a context vector for the image;
- and
- using the context vector to tag the image.
Dependencies are inferred from the claim text.