I-vectors for image classification

David C. Smith

doi:10.1117/12.2060207

23 September 2014 I-vectors for image classification

David C. Smith

Proceedings Volume 9217, Applications of Digital Image Processing XXXVII; 92170F (2014) https://doi.org/10.1117/12.2060207
Event: SPIE Optical Engineering + Applications, 2014, San Diego, California, United States

Abstract

Recent state-of-the-art work on speaker recognition and verification uses a simple factor analysis to derive a low-dimensional total variability space" which simultaneously captures speaker and channel variability. This approach simplified earlier work using joint factor analysis to separately model speaker and channel differences. Here we adapt this "i-vector" method to image classification by replacing speakers with image categories, voice cuts with images, and cepstral features with SURF local descriptors, and where the role of channel variability is attributed to differences in image backgrounds or lighting conditions. A Universal Gaussian mixture model (UGMM) is trained (unsupervised) on SURF descriptors extracted from a varied and extensive image corpus. Individual images are modeled by additively perturbing the supervector of stacked means of this UGMM by the product of a low-rank total variability matrix (TVM) and a normally distributed hidden random vector, X. The TVM is learned by applying an EM algorithm to maximize the sum of log-likelihoods of descriptors extracted from training images, where the likelihoods are computed with respect to the GMM obtained by perturbing the UGMM means via the TVM as above, and leaving UGMM covariances unchanged. Finally, the low-dimensional i-vector representation of an image is the expected value of the posterior distribution of X conditioned on the image's descriptors, and is computed via straightforward matrix manipulations involving the TVM and image-specific Baum-Welch statistics. We compare classification rates found with (i) i-vectors (ii) PCA (iii) Discriminant Attribute Projection (the last two trained on Gaussian MAP-adapted supervector image representations), and (iv) replacing the TVM with the matrix of dominant PCA eigenvectors before i-vector extraction.

Citation Download Citation

David C. Smith "I-vectors for image classification", Proc. SPIE 9217, Applications of Digital Image Processing XXXVII, 92170F (23 September 2014); https://doi.org/10.1117/12.2060207

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available