Active query-driven visual search using probabilistic bisection and convolutional neural networks

TL;DR

This paper introduces an efficient object detection method combining probabilistic bisection with CNN-based noisy oracle responses, significantly improving speed while maintaining accuracy in real-world face localization tasks.

Contribution

It proposes a novel query-driven visual search framework that leverages probabilistic bisection and CNNs for faster and accurate object localization.

Findings

Achieves the same lower bound on localization error as joint query design.

Speeds up face localization by at least an order of magnitude compared to sliding window.

Maintains high accuracy in real-world face localization tasks.

Abstract

We present a novel efficient object detection and localization framework based on the probabilistic bisection algorithm. A Convolutional Neural Network (CNN) is trained and used as a noisy oracle that provides answers to input query images. The responses along with error probability estimates obtained from the CNN are used to update beliefs on the object location along each dimension. We show that querying along each dimension achieves the same lower bound on localization error as the joint query design. Finally, we compare our approach to the traditional sliding window technique on a real world face localization task and show speed improvements by at least an order of magnitude while maintaining accurate localization.