Self-normalized linear tests

Authors:
Sachin Gangaputra;Donald Geman
Affiliations:
Dept. of Electrical and Computer Engineering, The Johns Hopkins University, Baltimore, MD;Dept. of Applied Mathematics and Statistics, The Johns Hopkins University, Baltimore, MD
Venue:
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Year:
2004

Citing 10
Cited 1

The nature of statistical learning theory

The nature of statistical learning theory
Neural Network-Based Face Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Example-Based Learning for View-Based Human Face Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
A computational model for visual selection

Neural Computation
An introduction to support Vector Machines: and other kernel-based learning methods

An introduction to support Vector Machines: and other kernel-based learning methods
Coarse-to-Fine Face Detection

International Journal of Computer Vision - Special issue on statistical and computational theories of vision: Part II
Training Support Vector Machines: an Application to Face Detection

CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Probabilistic Modeling of Local Appearance and Spatial Relationships for Object Recognition

CVPR '98 Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Perceptrons: An Introduction to Computational Geometry

Perceptrons: An Introduction to Computational Geometry

A Two-Level Approach Towards Semantic Colon Segmentation: Removing Extra-Colonic Findings

MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Making decisions based on a linear combination L of features is of course very common in pattern recognition. For distinguishing between two hypotheses or classes, the test is of the form sign(L - τ ) for some threshold τ. Due mainly to fixing τ , such tests are sensitive to changes in illumination and other variations in imaging conditions. We propose a special case, a "self-normalized linear test" (SNLT), hard-wired to be of the form sign(L1 - L2) with unit weights. The basic idea is to "normalize" L1, which involves the usual discriminating features, by L2, which is composed of non-discriminating features. For a rich variety of features (e.g., based directly on intensity differences), SNLTs are largely invariant to illumination and robust to unexpected background variations. Experiments in face detection are promising: they confirm the expected invariances and out-perform some previous results in a hierarchical framework.