Multi-task Image Classification via Collaborative, Hierarchical Spike-and-Slab Priors

TL;DR

This paper introduces a multi-task image classification method using collaborative, hierarchical spike-and-slab priors that leverage joint information from multiple views, improving recognition performance in low training scenarios.

Contribution

It extends class-specific spike-and-slab priors to multi-task settings, enabling effective multi-view face recognition with limited training data.

Findings

Improved multi-view face recognition accuracy.

Effective use of joint information from multiple camera views.

Enhanced performance in low training data scenarios.

Abstract

Promising results have been achieved in image classification problems by exploiting the discriminative power of sparse representations for classification (SRC). Recently, it has been shown that the use of \emph{class-specific} spike-and-slab priors in conjunction with the class-specific dictionaries from SRC is particularly effective in low training scenarios. As a logical extension, we build on this framework for multitask scenarios, wherein multiple representations of the same physical phenomena are available. We experimentally demonstrate the benefits of mining joint information from different camera views for multi-view face recognition.