Density-based clustering with fully-convolutional networks for crowd flow detection from drones

TL;DR

This paper introduces a novel fully-convolutional network approach for crowd flow detection from drone video sequences, enabling effective crowd clustering and movement tracking to analyze high-level crowd behavior.

Contribution

It presents a new deep learning-based method specifically designed for crowd flow detection in drone videos, addressing a previously unexplored research area.

Findings

Effective crowd clustering and movement tracking demonstrated

Proven on VisDrone challenge datasets with video sequences

Potential to advance high-level crowd behavior analysis from drones

Abstract

Crowd analysis from drones has attracted increasing attention in recent times due to the ease of use and affordable cost of these devices. However, how this technology can provide a solution to crowd flow detection is still an unexplored research question. To this end, we propose a crowd flow detection method for video sequences shot by a drone. The method is based on a fully-convolutional network that learns to perform crowd clustering in order to detect the centroids of crowd-dense areas and track their movement in consecutive frames. The proposed method proved effective and efficient when tested on the Crowd Counting datasets of the VisDrone challenge, characterized by video sequences rather than still images. The encouraging results show that the proposed method could open up new ways of analyzing high-level crowd behavior from drones.