MSCDP: Multi-step crowd density predictor in indoor environment

نویسندگان

چکیده

Monitoring and predicting crowd movements in indoor environments are of great importance management to prevent crushing trampling. Existing works mostly focused on individual trajectory forecasting a less crowded scene, or counting density estimation. Only very few predict the distribution. However, this study is failing realize multi-step prediction exploits only heatmaps modality ignores information complementation with corresponding video frames. Therefore, we motivated distribution multiple time steps facilitate long-term prediction. In paper, Multi-Step Crowd Density Predictor (MSCDP) fuse frame sequences heatmaps, proposed accurately forecast future heatmaps. To capture periodic movement features, optical flow context memory (LOFCM) module designed store learnable patterns. We conducted extensive experiments two real-world datasets. Evaluation results show that our MSCDP outperforms state-of-the-art baseline techniques variants terms various errors, demonstrating effectiveness each its key components

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Remediation of ozone pollution by ornamental plants in indoor environment

The indoor air quality is much more matter of concern as relative to ambient or outdoor air quality, especially in the context of human health.  However, very few studies have been reported for remediation of indoor ozone by plant species. The main objective of this study is to evaluate ozone deposition velocities and ozone removal effectiveness of three indoor ornamental plant species (Dra...

متن کامل

Capacity of Multi-antenna Array Systems in Indoor Wireless Environment

Previous studies have shown that a wireless system using n transmitting and n receiving antennas can achieve a capacity n times higher than a singleantenna system in an independent Rayleigh fading environment. In this paper, we explore the capacities of multiple-element antenna arrays (MEAs) in a more realistic propagation environment simulated via the WiSE ray-tracing tool. We impose an averag...

متن کامل

Airborne Crowd Density Estimation

This paper proposes a new method for estimating human crowd densities from aerial imagery. Applications benefiting from an accurate crowd monitoring system are mainly found in the security sector. Normally crowd density estimation is done through in-situ camera systems mounted on high locations although this is not appropriate in case of very large crowds with thousands of people. Using airborn...

متن کامل

Modeling Crowd Evacuation from Indoor Spaces

A crowd in an emergency situation located inside a public space, they will evacuate from that space. Crowd evacuation is a dynamic process which relies to a variety of environmental factors, as well as crowd specifications itself. To help people out of danger, rapidly and safety, we need to have a previous plan for any public space, which is obtained by collecting information of crowd evacuatio...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Neurocomputing

سال: 2023

ISSN: ['0925-2312', '1872-8286']

DOI: https://doi.org/10.1016/j.neucom.2023.126296