Method, system, and device for learned invariant feature transform for computer images

Pascal Fua; Vincent Lepetit; Eduard  Fortuny Trulls; Kwang Moo Yi

Method, system, and device for learned invariant feature transform for computer images

Pascal Fua (Inventor), Vincent Lepetit (Inventor), Eduard Fortuny Trulls (Inventor), Kwang Moo Yi (Inventor)

Institute of Computer Graphics and Vision (7100)

Research output: Patent

Abstract

A method for training a feature detector of an image processing device, including the steps of detecting features in the image to generate a score map, computing a center of mass on the score map to generate a location, extracting a patch from the image at the location by a first spatial transformer, estimating an orientation of the patch, rotating the patch in accordance with the patch orientation with a second spatial transformer, and describing the rotated patch to create a description vector.

Original language	English
Publication status	Published - 4 Feb 2020

Cite this

@misc{7ab6091f0b714cfaa16456683fb1ca38,

title = "Method, system, and device for learned invariant feature transform for computer images",

abstract = "A method for training a feature detector of an image processing device, including the steps of detecting features in the image to generate a score map, computing a center of mass on the score map to generate a location, extracting a patch from the image at the location by a first spatial transformer, estimating an orientation of the patch, rotating the patch in accordance with the patch orientation with a second spatial transformer, and describing the rotated patch to create a description vector.",

author = "Pascal Fua and Vincent Lepetit and {Fortuny Trulls}, Eduard and Yi, {Kwang Moo}",

year = "2020",

month = feb,

day = "4",

language = "English",

type = "Patent",

}

TY - PAT

T1 - Method, system, and device for learned invariant feature transform for computer images

AU - Fua, Pascal

AU - Lepetit, Vincent

AU - Fortuny Trulls, Eduard

AU - Yi, Kwang Moo

PY - 2020/2/4

Y1 - 2020/2/4

N2 - A method for training a feature detector of an image processing device, including the steps of detecting features in the image to generate a score map, computing a center of mass on the score map to generate a location, extracting a patch from the image at the location by a first spatial transformer, estimating an orientation of the patch, rotating the patch in accordance with the patch orientation with a second spatial transformer, and describing the rotated patch to create a description vector.

AB - A method for training a feature detector of an image processing device, including the steps of detecting features in the image to generate a score map, computing a center of mass on the score map to generate a location, extracting a patch from the image at the location by a first spatial transformer, estimating an orientation of the patch, rotating the patch in accordance with the patch orientation with a second spatial transformer, and describing the rotated patch to create a description vector.

M3 - Patent

ER -