Real-Time Deep Learning Method for Abandoned Luggage Detection in Video

TL;DR

This paper presents a real-time deep learning system for detecting abandoned luggage in surveillance videos, combining background subtraction and CNN-based recognition, with improved performance over existing methods.

Contribution

It introduces a two-stage approach integrating background subtraction and CNNs for real-time abandoned luggage detection in video footage.

Findings

Outperforms a strong CNN baseline in accuracy

Uses diverse training data including internet images and synthetic scenes

Achieves real-time detection suitable for surveillance applications

Abstract

Recent terrorist attacks in major cities around the world have brought many casualties among innocent citizens. One potential threat is represented by abandoned luggage items (that could contain bombs or biological warfare) in public areas. In this paper, we describe an approach for real-time automatic detection of abandoned luggage in video captured by surveillance cameras. The approach is comprised of two stages: (i) static object detection based on background subtraction and motion estimation and (ii) abandoned luggage recognition based on a cascade of convolutional neural networks (CNN). To train our neural networks we provide two types of examples: images collected from the Internet and realistic examples generated by imposing various suitcases and bags over the scene's background. We present empirical results demonstrating that our approach yields better performance than a strong CNN baseline method.