CrossCount: A Deep Learning System for Device-Free Human Counting Using WiFi

Ibrahim, Osama Talaat; Gomaa, Walid; Youssef, Moustafa

doi:10.1109/jsen.2019.2928502

Cited by 56 publications

(16 citation statements)

References 22 publications

(38 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Only 9 people were considered to validate the effectiveness of this technique. An interesting development is presented in [17], where WiFi linkage blockage patterns are classified using LSTM for crowd counting. It remains to be seen whether the system can perform equally well in outdoor environments with large and dense crowd.…”

Section: B Related Workmentioning

confidence: 99%

Crowd Forecasting Based on WiFi Sensors and LSTM Neural Networks

Singh

Determe

Horlin

et al. 2020

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

To ensure effective management and security in large scale public events, it is imperative for the event organizers to be aware of potentially critical crowd densities. This paper, therefore, presents a solution to the above problem in terms of WiFi based crowd counting and LSTM neural network based forecasting. Monitoring of an actual event organized in Brussels has been described, wherein crowd counts are obtained using WiFi sensors in a privacy-preserved manner. The time-stamped crowd counts are used to develop univariate time-series, which are in-turn utilized for forecasting. Five different LSTM models are utilized for crowd time-series forecasting and analyzed for their suitability. A random walk model is used as reference for performance assessment. Among different LSTM models, Convolutional LSTM delivered the best performance. Overall results and analysis show that the developed system is suitable for crowd monitoring.

show abstract

Section: B Related Workmentioning

confidence: 99%

Crowd Forecasting Based on WiFi Sensors and LSTM Neural Networks

Singh

Determe

Horlin

et al. 2020

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

show abstract

“…The accuracy is good and the sensors are less invasive, but array sensors are still quite expensive. Ultrasonic arrays [18], [19] and WiFi-based [20], [21] device-free passive approaches are the least invasive, but potentially have lower people counting accuracy.…”

Section: B Past Approaches To People Countingmentioning

confidence: 99%

Single-Pixel Thermopile Infrared Sensing for People Counting

Hagenaars

Pandharipande

Murthy³

et al. 2021

IEEE Sensors J.

View full text Add to dashboard Cite

People-counting data can be used in buildingcontrol systems to improve comfort and in space management applications to optimize building space. In this work, we consider a combi-sensor with a single-pixel thermopile and passive infrared sensor for people counting. We first develop a thermopile signal model for object temperature measurements under multiple people occupancy. We then propose a people counting method based on: cumulative sum (CUSUM) change detection in the object temperature signal, forming a people count estimate using likelihoods of differential mean temperature in detected changes, and decision fusion with an infrared vacancy sensor. The proposed method is evaluated with data generated using the developed signal model as well as experimental data from a cell office/meeting room environment. We obtain an average counting error of 0.11 and 0.19 for 90% of the instants respectively when considering 15 minute windows for simulated and experimental datasets.

show abstract

“…Nevertheless, for the sake of completion, we shall review this category of work next. These counting methods can be broadly divided into two categories: model-based methods [6,7,37], and learning-based methods [9,13,15,24,39]. In model-based methods, the counting is based on a mathematical modeling of the received WiFi signal based on people's motion.…”

Section: Counting a Mobile Crowdmentioning

confidence: 99%

Counting a stationary crowd using off-the-shelf wifi

Korany

Mostofi

2021

Proceedings of the 19th Annual International Conference on Mobile Systems, Applications, and Services

View full text Add to dashboard Cite

In this paper, we are interested in the problem of counting a crowd of stationary people (i.e., seated) using a pair of WiFi transceivers. While the people in the crowd are stationary, i.e. with no major body motion except breathing, people do not stay still for a long period of time and frequently engage in small in-place body motions called fidgets (e.g., adjusting their seating position, crossing their legs, checking their phones, etc). In this paper, we propose that the aggregate natural fidgeting and in-place motions of a stationary crowd carry crucial information on the crowd count. We then mathematically characterize the Probability Distribution Function (PDF) of the crowd fidgeting and silent periods (which we can extract from the received WiFi signal) and show their dependency on the total number of people in the area. In developing our mathematical models, we show how our problem of interest resembles a severaldecade-old / /∞ queuing theory problem, which allows us to borrow mathematical tools from the literature on / /∞ queues. We extensively validate our proposed approach with a total of 47 experiments in four different environments (including through-wall settings), in which up to and including = 10 people are seated. We further test our system in different scenarios, and with different activities, representing various engagement levels of the crowd, such as attending a lecture, watching a movie, and reading. Moreover, we test our proposed system with different number of people seated in several different configurations. Our evaluation results show that our proposed approach achieves a very high counting accuracy, with the estimated number of people being only 0 or 1 off from the true number 96.3% of the time in non-through-wall settings, and 90% of the time in through-wall settings. Our results show the potential of our proposed framework for crowd counting in real-world scenarios. CCS CONCEPTS• Human-centered computing → Ubiquitous and mobile computing; • Mathematics of computing → Queueing theory; • Computer systems organization → Sensor networks.

show abstract

CrossCount: A Deep Learning System for Device-Free Human Counting Using WiFi

Cited by 56 publications

References 22 publications

Crowd Forecasting Based on WiFi Sensors and LSTM Neural Networks

Crowd Forecasting Based on WiFi Sensors and LSTM Neural Networks

Single-Pixel Thermopile Infrared Sensing for People Counting

Counting a stationary crowd using off-the-shelf wifi

Contact Info

Product

Resources

About