Learning apparatus, leaning method and process

Abstract

The present invention discloses a learning apparatus, a learning method and a process. The learning apparatus includes a learning section which learns, according as a learning image used for learning a discriminator for discriminating whether a predetermined discrimination target is present in an image is designated from a plurality of sample images by a user, the discriminator using a random feature amount including a dimension feature amount randomly selected from a plurality of dimension feature amounts included in an image feature amount indicating features of the learning image.

Description

Technical field [0001] The present invention relates to a learning device, a learning method, and a program, and more specifically, to a learning device, a learning method, and a program suitable for use in the case of training a discriminator, which is used to learn images based on a small amount of , To identify whether there is a predetermined identification target in the image. Background technique [0002] In the prior art, an image classification method is proposed for classifying multiple images into categories corresponding to the subject of the image, and generating image clusters including classified images for each category. [0003] For example, in this image classification method, a discriminator that discriminates whether a predetermined discrimination target (for example, a human face) exists in an image is used to discriminate whether a predetermined discrimination target exists in each of a plurality of images. [0004] In addition, according to the authentication r...

AI Technical Summary

Problems solved by technology

[0017] However, even in the method using bagging, although the number of learning images is small, in the case of using image feature quantities expressed as learning images having vectors of hundreds to thousands of dimensions, overshooting is expected to occur. study

Images