Cognitive integrated image classification and annotation

Patent metadata

Field	Value
Publication number	US-2018342093-A1
Application number	US-201715840282-A
Country	US
Kind code	A1
Filing date	Dec 13, 2017
Priority date	May 26, 2017
Publication date	Nov 29, 2018
Grant date	—

Abstract

A method for classifying and annotating an image includes: receiving, by a computer device and from a user interface, an input of an image; generating an annotation of the image, by the computer device, by passing the image to plural separate pipelines and tag libraries, wherein the plural separate pipelines and tag libraries include: a pipeline configured to classify and tag objects in the image; and a pipeline configured to tag kinematic aspects of the objects in the image; and outputting, by the computer device, the annotation to the user interface.

First claim

What is claimed is:

1 . A method for classifying and annotating an image, comprising:
receiving, by a computer device and from a user interface, an input of an image; generating an annotation of the image, by the computer device, by passing the image to plural separate pipelines and tag libraries, wherein the plural separate pipelines and tag libraries comprise: a pipeline configured to classify and tag objects in the image; and a pipeline configured to tag kinematic aspects of the objects in the image; and outputting, by the computer device, the annotation to the user interface.

2 . The method of claim 1 , wherein the plural separate pipelines and tag libraries comprise a pipeline configured to classify and tag an aggregation of the objects in the image.

3 . The method of claim 1 , wherein the plural separate pipelines and tag libraries comprise a pipeline configured to classify and tag an aggregation of the kinematic aspects of the objects in the image.

4 . The method of claim 1 , wherein the plural separate pipelines and tag libraries comprise a pipeline configured to classify and tag a situation in the image.

5 . The method of claim 1 , wherein the pipeline configured to classify and tag objects in the image uses predefined object templates to classify the objects.

6 . The method of claim 1 , wherein the pipeline configured to tag kinematic aspects of the objects in the image uses predefined kinematic templates to classify the kinematic aspects.

7 . The method of claim 1 , further comprising passing the image to each of the plural separate pipelines and tag libraries in a predefined order.

8 . The method of claim 1 , further comprising:
obtaining insights about one or more of the objects in the image from a big data platform; and adjusting one or more object tags of the image based on the insights.

9 . The method of claim 8 , wherein the adjusting the one or more object tags comprises replacing a generic object tag with one of a name, a relationship, and an age descriptor.

10 . The method of claim 1 , wherein the image comprises a sequence of plural images, and further comprising:
performing the generating an annotation for each one of the plural images; and eliminating redundant tags from consecutive ones of the plural images.

Assignees

• IBM

Inventors

• Choo Kristina Y
• Hodge Rashida A
• Ramachandran Krishnan K
• Sivakumar Gandhi

Classifications

• G06V20/70 Primary

  Labelling scene content, e.g. deriving syntactic or semantic representations · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06T11/60 Primary

  Creating or editing images; Combining images with text · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06K9/6267

  Physics · mapped topic