LEARNING INVARIANT COLOR FEATURES WITH SPARSE TOPOGRAPHIC RESTRICTED BOLTZMANN MACHINES

被引：0

作者：

Goh, Hanlin ^{[1
]}

Kusmierz, Lukasz ^{[1
]}

Lim, Joo-Hwee ^{[1
]}

Thome, Nicolas ^{[2
]}

Cord, Matthieu ^{[2
]}

机构：

[1] ASTAR, Inst Infocomm Res, Singapore, Singapore

[2] UPMC, Sorbonne Univ, Lab Informat Paris 6, Paris, France

来源：

2011 18TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP) | 2011年

关键词：

Unsupervised feature learning; invariant features; sparse coding; topographic coding; color features;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Our objective is to learn invariant color features directly from data via unsupervised learning. In this paper, we introduce a method to regularize restricted Boltzmann machines during training to obtain features that are sparse and topographically organized. Upon analysis, the features learned are Gabor-like and demonstrate a coding of orientation, spatial position, frequency and color that vary smoothly with the topography of the feature map. There is also differentiation between monochrome and color filters, with some exhibiting color-opponent properties. We also found that the learned representation is more invariant to affine image transformations and changes in illumination color.

引用

页码：1241 / 1244

页数：4

共 50 条

[1] Learning Restricted Boltzmann Machines with Sparse Latent Variables
Bresler, Guy
Buhai, Rares-Darius
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
[2] LEARNING SPAM FEATURES USING RESTRICTED BOLTZMANN MACHINES
da Silva, Luis Alexandre
Pontara da Costa, Kelton Augusto
Ribeiro, Patricia Bellin
de Rosa, Gustavo Henrique
Papa, Joao Paulo
IADIS-INTERNATIONAL JOURNAL ON COMPUTER SCIENCE AND INFORMATION SYSTEMS, 2016, 11 (01): : 99 - 114
[3] Learning Informative Features from Restricted Boltzmann Machines
Tomczak, Jakub M.
NEURAL PROCESSING LETTERS, 2016, 44 (03) : 735 - 750
[4] Learning Informative Features from Restricted Boltzmann Machines
Jakub M. Tomczak
Neural Processing Letters, 2016, 44 : 735 - 750
[5] Learning Invariant Features Using Subspace Restricted Boltzmann Machine
Jakub M. Tomczak
Adam Gonczarek
Neural Processing Letters, 2017, 45 : 173 - 182
[6] Learning Invariant Features Using Subspace Restricted Boltzmann Machine
Tomczak, Jakub M.
Gonczarek, Adam
NEURAL PROCESSING LETTERS, 2017, 45 (01) : 173 - 182
[7] Stacks of Convolutional Restricted Boltzmann Machines for Shift-Invariant Feature Learning
Norouzi, Mohammad
Ranjbar, Mani
Mori, Greg
CVPR: 2009 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, VOLS 1-4, 2009, : 2727 - 2734
[8] Weakly supervised learning with denoising restricted boltzmann machines for extracting features
Yang, Jie
Sun, Ya-Dong
Zhang, Liang-Jun
Liu, Hai-Bo
Tien Tzu Hsueh Pao/Acta Electronica Sinica, 2014, 42 (12): : 2365 - 2370
[9] Biasing restricted Boltzmann machines using Gaussian filters to learn invariant visual features
Yogeswaran, Arjun
Payeur, Pierre
2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016, 2017,
[10] Biasing Restricted Boltzmann Machines using Gaussian Filters to Learn Invariant Visual Features
Yogeswaran, Arjun
Payeur, Pierre
PROCEEDINGS OF 2016 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (SSCI), 2016,

← 1 2 3 4 5 →