Joint calibration of Ensemble of Exemplar SVMs

TL;DR

This paper introduces a joint calibration method for Ensemble of Exemplar SVMs that optimizes ensemble performance collectively, leading to improved classification and detection results on benchmark datasets.

Contribution

It proposes a novel joint calibration approach formulated as a constrained optimization problem with an efficient algorithm, outperforming independent calibration methods.

Findings

Outperforms independent calibration on classification tasks

Leads to better object detection performance

Demonstrated on ILSVRC 2014 and PASCAL VOC 2007 datasets

Abstract

We present a method for calibrating the Ensemble of Exemplar SVMs model. Unlike the standard approach, which calibrates each SVM independently, our method optimizes their joint performance as an ensemble. We formulate joint calibration as a constrained optimization problem and devise an efficient optimization algorithm to find its global optimum. The algorithm dynamically discards parts of the solution space that cannot contain the optimum early on, making the optimization computationally feasible. We experiment with EE-SVM trained on state-of-the-art CNN descriptors. Results on the ILSVRC 2014 and PASCAL VOC 2007 datasets show that (i) our joint calibration procedure outperforms independent calibration on the task of classifying windows as belonging to an object class or not; and (ii) this improved window classifier leads to better performance on the object detection task.