CMU VASC seminar: Building Classification Cascades for Visual Identification from One Example

Andras Ferencz
UC Berkeley (now at Mobileye Vision Technologies)

Abstract:
I will describe our effort to solve the problem of object identification (OID), which is specialized recognition where the category is known (for example cars or faces) and the algorithm recognizes an object's exact identity. Two special challenges characterize OID: 1. Inter-class variation is often small (many cars look alike) and may be dwarfed by illumination or pose changes; 2. There may be many classes but few or just one positive "training" examples per class.

Due to (1), a solution must locate possibly subtle object-specific salient features (a door handle) while avoiding distracting ones (a specular highlights). However, (2) rules out direct techniques of feature selection. I will describe an on-line algorithm that takes one query image from a known category and builds an efficient "same" vs. "different" classification cascade by predicting the most discriminative feature set for that object. Our method not only estimates the saliency and scoring function for each candidate feature, but also models the dependency between features, building an ordered feature sequence unique to a specific query image, maximizing cumulative information content. Learned stopping thresholds make the classifier very efficient. To make this possible, category-specific characteristics are learned automatically in an off-line training procedure from labeled image pairs of the category, without prior knowledge about the category.

Andras Ferencz, Erik Learned-Miller, Jitendra Malik. Building a Classification Cascade for Visual Identification from One Example.
Draft: submitted to ICCV 2005. PDF, Project Page