Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds

TL;DR

This paper introduces a unified deep learning approach for counting, density map estimation, and localization in dense crowd images, supported by a new large-scale dataset and outperforming existing methods.

Contribution

A novel decomposable loss function for deep CNNs that jointly addresses counting, density estimation, and localization in dense crowds, along with a new dataset.

Findings

Significantly outperforms existing methods on the new dataset.

Introduces the UCF-QNRF dataset with 1.25 million annotations.

Demonstrates the effectiveness of joint optimization for crowd analysis tasks.

Abstract

With multiple crowd gatherings of millions of people every year in events ranging from pilgrimages to protests, concerts to marathons, and festivals to funerals; visual crowd analysis is emerging as a new frontier in computer vision. In particular, counting in highly dense crowds is a challenging problem with far-reaching applicability in crowd safety and management, as well as gauging political significance of protests and demonstrations. In this paper, we propose a novel approach that simultaneously solves the problems of counting, density map estimation and localization of people in a given dense crowd image. Our formulation is based on an important observation that the three problems are inherently related to each other making the loss function for optimizing a deep CNN decomposable. Since localization requires high-quality images and annotations, we introduce UCF-QNRF dataset that overcomes the shortcomings of previous datasets, and contains 1.25 million humans manually marked with dot annotations. Finally, we present evaluation measures and comparison with recent deep CNN networks, including those developed specifically for crowd counting. Our approach significantly outperforms state-of-the-art on the new dataset, which is the most challenging dataset with the largest number of crowd annotations in the most diverse set of scenes.