Counting Everyday Objects in Everyday Scenes

TL;DR

This paper introduces a novel counting model inspired by human subitizing, designed to handle the variability in natural scenes, and demonstrates its effectiveness on standard datasets and its potential to enhance object detection and VQA tasks.

Contribution

The paper presents a dedicated counting approach that leverages subitizing principles and contextual information to improve counting accuracy in natural images.

Findings

Improved counting accuracy on PASCAL VOC 2007 and COCO datasets.

Counting enhances object detection performance.

Counting methods aid in answering 'how many?' questions in VQA.

Abstract

We are interested in counting the number of instances of object classes in natural, everyday images. Previous counting approaches tackle the problem in restricted domains such as counting pedestrians in surveillance videos. Counts can also be estimated from outputs of other vision tasks like object detection. In this work, we build dedicated models for counting designed to tackle the large variance in counts, appearances, and scales of objects found in natural scenes. Our approach is inspired by the phenomenon of subitizing - the ability of humans to make quick assessments of counts given a perceptual signal, for small count values. Given a natural scene, we employ a divide and conquer strategy while incorporating context across the scene to adapt the subitizing idea to counting. Our approach offers consistent improvements over numerous baseline approaches for counting on the PASCAL VOC 2007 and COCO datasets. Subsequently, we study how counting can be used to improve object detection. We then show a proof of concept application of our counting methods to the task of Visual Question Answering, by studying the `how many?' questions in the VQA and COCO-QA datasets.