arrow
Return

Learning Deep Hierarchical Visual Feature Coding

delete2014-12-01
delete82
PRE
AI
H
Hanlin Goh *
N
Nicolas Thome
M
Matthieu Cord
J
Joo‐Hwee Lim
DOI:10.1109/TNNLS.2014.2307532delete
deleteOriginal
deleteShare
deleteSave
Abstract

Abstract

En 中文
In this paper, we propose a hybrid architecture that combines the image modeling strengths of the bag of words framework with the representational power and adaptability of learning deep architectures. Local gradient-based descriptors, such as SIFT, are encoded via a hierarchical coding scheme composed of spatial aggregating restricted Boltzmann machines (RBM). For each coding layer, we regularize the RBM by encouraging representations to fit both sparse and selective distributions. Supervised fine-tuning is used to enhance the quality of the visual representation for the categorization task. We performed a thorough experimental evaluation using three image categorization data sets. The hierarchical coding scheme achieved competitive categorization accuracies of 79.7% and 86.4% on the Caltech-101 and 15-Scenes data sets, respectively. The visual representations learned are compact and the model's inference is fast, as compared with sparse coding methods. The low-level representations of descriptors that were learned using this method result in generic features that we empirically found to be transferrable between different image data sets. Further analysis reveal the significance of supervised fine-tuning when the architecture has two layers of representations as opposed to a single layer.
Keywords:
Bag-of-words (BoW) framework
computer vision
deep learning
dictionary learning
hierarchical visual architecture
image categorization
restricted Boltzmann machine (RBM)
sparse feature coding
transfer learning

Journal

IEEE Transactions on Neural Networks and Learning Systems cover
IEEE Transactions on Neural Networks and Learning Systems
IF:
8.9
Papers:
7.8K
Citations:
9.0W

Organization

S
sorbonne-universite
Scholars:
6
Papers: 4
Citations: 0
A
agency for science technology & research (a*star)
Scholars:
2.2W
Papers: 1.9W
Citations: 3.4W
Cited Papers

Cited Papers

A novel endoesophageal magnetic device to prevent gastroesophageal reflux
err2008-12-31
err0
PREAI
errMauro Bortolotti; Annamaria Grandis; Giosuè Mazzero
errShare
errSave
Primary Headache Disorders
err2013-07-01
err0
PREAI
errRafael Benoliel; Eli Eliav
errShare
errSave
Diseases and Molecular Diagnostics: A Step Closer to Precision Medicine
err2017-08-22
err0
errOAAI
errShailendra Dwivedi; Purvi Purohit; Radhieka Misra; Puneet Pareek; Apul Goel; Sanjay Khattri; Kamlesh Kumar Pant; Sanjeev Misra; Praveen Sharma
errShare
errSave
Crystalline‐State Reaction with Allosteric Effect in Spin‐Crossover, Interpenetrated Networks with Magnetic and Optical Bistability
err2003-08-13
err0
PREAI
errVirginie Niel; Amber L. Thompson; M. Carmen Muñoz; Ana Galet; Andrés E. Goeta; José A. Real
errShare
errSave
errShare
errSave
researcher View more