Compression Techniques for Deep Fisher Vectors

This paper investigates the use of efficient compression tec hniques for Fisher vectors derived from deep architectures such as restricted Boltzmann machine (RBM). Fi sher representations have recently created a surge of interest by proving their worth for large scale object rec ognition and retrieval problems due to the intrinsic properties that make them unique from the conventional bag o f visual words (BoW) features, however they suffer from the problem of large dimensionality. This paper rovides empirical evidence along with visualisations to explore which of the feature normalisation and stat e of the art compression techniques is well suited for deep Fisher vectors, making them amenable for large scale visual retrieval with reduced memory footprint. We further show that the compressed Fisher vectors give impr essive classification results even with costless linear classifiers like k-nearest neighbour.

[1]  Luc Van Gool,et al.  The Pascal Visual Object Classes Challenge: A Retrospective , 2014, International Journal of Computer Vision.

[2]  Jianfei Cai,et al.  Compact Representation for Image Classification: To Choose or to Compress? , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[3]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[4]  Mahesan Niranjan,et al.  Inducing discrimination in biologically inspired models of visual scene recognition , 2013, 2013 IEEE International Workshop on Machine Learning for Signal Processing (MLSP).

[5]  John Shawe-Taylor,et al.  Improving "bag-of-keypoints" image categorisation: Generative Models and PDF-Kernels , 2005 .

[6]  Laurens van der Maaten,et al.  Learning a Parametric Embedding by Preserving Local Structure , 2009, AISTATS.

[7]  Gabriela Csurka,et al.  Adapted Vocabularies for Generic Visual Categorization , 2006, ECCV.

[8]  Andrew Zisserman,et al.  The devil is in the details: an evaluation of recent feature encoding methods , 2011, BMVC.

[9]  Gabriela Csurka,et al.  Visual categorization with bags of keypoints , 2002, eccv 2004.

[10]  Laurens van der Maaten,et al.  Learning Discriminative Fisher Kernels , 2011, ICML.

[11]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[12]  Florent Perronnin,et al.  Fisher vectors meet Neural Networks: A hybrid classification architecture , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  Geoffrey E. Hinton.  A Practical Guide to Training Restricted Boltzmann Machines , 2012, Neural Networks: Tricks of the Trade.

[14]  Thomas Mensink,et al.  Compressed Fisher Vectors for Large-Scale Image Classification , 2013 .

[15]  Antonio Torralba,et al.  Spectral Hashing , 2008, NIPS.

[16]  S. Mohamed,et al.  Statistical Normalization and Back Propagation for Classification , 2022 .

[17]  Jean Ponce,et al.  Learning mid-level features for recognition , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[18]  Cordelia Schmid,et al.  Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[19]  Karl Pearson F.R.S..  LIII. On lines and planes of closest fit to systems of points in space , 1901 .

[20]  Thomas Mensink,et al.  Improving the Fisher Kernel for Large-Scale Image Classification , 2010, ECCV.

[21]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[22]  Florent Perronnin,et al.  Fisher Kernels on Visual Vocabularies for Image Categorization , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[23]  David Haussler,et al.  Exploiting Generative Models in Discriminative Classifiers , 1998, NIPS.