Patent · US Active

Neural network patch aggregation and statistics

US9996768B2 · kind B2 · utility

5Cited by
9References
25Claims
0Family size

Assignee

Inventors

Key dates

Filing dateNov 19, 2014
Grant dateJun 12, 2018
Priority date
Expiry dateOct 10, 2036

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/048
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Neural network patch aggregation and statistical techniques are described. In one or more implementations, patches are generated from an image, e.g., randomly, and used to train a neural network. An aggregation of outputs of patches processed by the neural network may be used to label an image using an image descriptor, such as to label aesthetics of the image, classify the image, and so on. In another example, the patches may be used by the neural network to calculate statistics describing the patches, such as to describe statistics such as minimum, maximum, median, and average of activations of image characteristics of the individual patches. These statistics may also be used to support a variety of functionality, such as to label the image as described above.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.