Single Image Object Counting and Localizing using Active-Learning

TL;DR

This paper introduces an active-learning based method for counting and localizing objects in a single image without pre-trained classifiers, optimizing label efficiency and accuracy through minimal user interaction.

Contribution

It presents a novel active-learning approach that trains a CNN on few labels from a single image, improving counting and localization accuracy with minimal user effort.

Findings

Achieves state-of-the-art accuracy in counting and localization

Reduces user labeling effort compared to existing methods

Demonstrates robustness across diverse image conditions

Abstract

The need to count and localize repeating objects in an image arises in different scenarios, such as biological microscopy studies, production lines inspection, and surveillance recordings analysis. The use of supervised Convoutional Neural Networks (CNNs) achieves accurate object detection when trained over large class-specific datasets. The labeling effort in this approach does not pay-off when the counting is required over few images of a unique object class. We present a new method for counting and localizing repeating objects in single-image scenarios, assuming no pre-trained classifier is available. Our method trains a CNN over a small set of labels carefully collected from the input image in few active-learning iterations. At each iteration, the latent space of the network is analyzed to extract a minimal number of user-queries that strives to both sample the in-class manifold as thoroughly as possible as well as avoid redundant labels. Compared with existing user-assisted counting methods, our active-learning iterations achieve state-of-the-art performance in terms of counting and localizing accuracy, number of user mouse clicks, and running-time. This evaluation was performed through a large user study over a wide range of image classes with diverse conditions of illumination and occlusions.