Learning Person-Specific Representations From Faces in the Wild
- 19 September 2014
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 9 (12) , 2089-2099
Abstract
Humans are natural face recognition experts, far out-performing current automated face recognition algorithms, especially in naturalistic, “in the wild” settings. However, a striking feature of human face recognition is that we are dramatically better at recognizing highly familiar faces, presumably because we can leverage large amounts of past experience with the appearance of an individual to aid future recognition. Meanwhile, the analogous situation in automated face recognition, where a large number of training examples of an individual are available, has been largely underexplored, in spite of the increasing relevance of this setting in the age of social media. Inspired by these observations, we propose to explicitly learn enhanced face representations on a per-individual basis, and we present two methods enabling this approach. By learning and operating within person-specific representations, we are able to significantly outperform the previous state-of-the-art on PubFig83, a challenging benchmark for familiar face recognition in the wild, using a novel method for learning representations in deep visual hierarchies. We suggest that such person-specific representations aid recognition by introducing an intermediate form of regularization to the problem.Keywords
Funding Information
- Fundação de Amparo à Pesquisa do Estado de São Paulo (2010/00994-8, 2010/05647-4, 2013/11359-0)
- National Council for Scientific and Technological Development, Brasília, Brazil (303673/2010-9, 304352/2012-8, 477662/2013-7, 479070/2013-0)
- Microsoft Research, Redmond, WA, USA
- Rowland Institute at Harvard, Cambridge, MA, USA
This publication has 42 references indexed in Scilit:
- Face Identification Using Large Feature SetsIEEE Transactions on Image Processing, 2011
- LIBSVMACM Transactions on Intelligent Systems and Technology, 2011
- Vehicle Detection Using Partial Least SquaresIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Partial least squares regression and projection on latent structure regression (PLS Regression)WIREs Computational Statistics, 2010
- Human detection using partial least squares analysis2009
- Overview and Recent Advances in Partial Least Squares2006
- Tumor classification by partial least squares using microarray gene expression dataBioinformatics, 2002
- Random projection in dimensionality reduction2001
- Interactive variable selection (IVS) for PLS. Part II: Chemical applicationsJournal of Chemometrics, 1995
- Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in positionBiological Cybernetics, 1980