Learning Transformation Invariance for Object Recognition

Jens Hocke; Thomas Martinetz

Learning Transformation Invariance for Object Recognition

Institute of Neuro- and Bioinformatics

Abstract

Based on Tomaso Poggio’sM-theory, we propose a method to learn transformation invariant representations. Using an artificial dataset, we demonstrate that our supervised method learns invariance to shifts, and on the MNIST data we show first results for learning the unknown transformations underlying handwritten digits.

Original language	English
Title of host publication	Workshop New Challenges in Neural Computation 2014
Editors	Barbara Hammer, Thomas Villmann
Number of pages	6
Volume	02
Publication date	02.09.2014
Pages	20-25
Publication status	Published - 02.09.2014
Event	36th German Conference on Pattern Recognition - Münster, Germany Duration: 02.09.2014 → 05.09.2014 http://resources.mpi-inf.mpg.de/conferences/dagm/2014/workshops.html

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

http://www.inb.uni-luebeck.de/publications/pdfs/HoMa14b.pdf

Cite this

@inproceedings{febf23ffcddd437d96cbcbdd5ed1d7bd,

title = "Learning Transformation Invariance for Object Recognition",

abstract = "Based on Tomaso Poggio{\textquoteright}sM-theory, we propose a method to learn transformation invariant representations. Using an artificial dataset, we demonstrate that our supervised method learns invariance to shifts, and on the MNIST data we show first results for learning the unknown transformations underlying handwritten digits.",

author = "Jens Hocke and Thomas Martinetz",

year = "2014",

month = sep,

day = "2",

language = "English",

volume = "02",

series = "Machine Learning Reports",

pages = "20--25",

editor = "Barbara Hammer and Thomas Villmann",

booktitle = "Workshop New Challenges in Neural Computation 2014",

note = "36th German Conference on Pattern Recognition ; Conference date: 02-09-2014 Through 05-09-2014",

url = "http://resources.mpi-inf.mpg.de/conferences/dagm/2014/workshops.html",

}

TY - GEN

T1 - Learning Transformation Invariance for Object Recognition

AU - Hocke, Jens

AU - Martinetz, Thomas

PY - 2014/9/2

Y1 - 2014/9/2

N2 - Based on Tomaso Poggio’sM-theory, we propose a method to learn transformation invariant representations. Using an artificial dataset, we demonstrate that our supervised method learns invariance to shifts, and on the MNIST data we show first results for learning the unknown transformations underlying handwritten digits.

AB - Based on Tomaso Poggio’sM-theory, we propose a method to learn transformation invariant representations. Using an artificial dataset, we demonstrate that our supervised method learns invariance to shifts, and on the MNIST data we show first results for learning the unknown transformations underlying handwritten digits.

UR - https://www.techfak.uni-bielefeld.de/~fschleif/mlr/mlr.html

M3 - Conference contribution

VL - 02

T3 - Machine Learning Reports

SP - 20

EP - 25

BT - Workshop New Challenges in Neural Computation 2014

A2 - Hammer, Barbara

A2 - Villmann, Thomas

T2 - 36th German Conference on Pattern Recognition

Y2 - 2 September 2014 through 5 September 2014

ER -

Learning Transformation Invariance for Object Recognition

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this