MiCrowd: Vision-Based Deep Crowd Counting on MCU.

Sensors (Basel)

Graduate School of Data Science, Seoul National University, Seoul 08826, Republic of Korea.

Published: March 2023

Microcontrollers (MCUs) have been deployed on numerous IoT devices due to their compact sizes and low costs. MCUs are capable of capturing sensor data and processing them. However, due to their low computational power, applications processing sensor data with deep neural networks (DNNs) have been limited. In this paper, we propose , a floating population measurement system with a tiny DNNs running on MCUs since the data have essential value in urban planning and business. Moreover, addresses the following important challenges: (1) privacy issues, (2) communication costs, and (3) extreme resource constraints on MCUs. To tackle those challenges, we designed a lightweight crowd-counting deep neural network, named MiCrowdNet, which enables . In addition, our dataset is carefully chosen and completely re-labeled to train MiCrowdNet for counting people from an . Experiments show the effectiveness of MiCrowdNet and our relabeled dataset for accurate on-device crowd counting.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10098830PMC
http://dx.doi.org/10.3390/s23073586DOI Listing

Publication Analysis

Top Keywords

crowd counting
8
sensor data
8
deep neural
8
microwd vision-based
4
vision-based deep
4
deep crowd
4
counting mcu
4
mcu microcontrollers
4
mcus
4
microcontrollers mcus
4

Similar Publications

Multiscale regional calibration network for crowd counting.

Sci Rep

January 2025

College of Computer Science and Software Engineering, Hohai University, Nanjing, 211100, China.

Crowd counting aims to estimate the number, density, and distribution of crowds in an image. While CNN-based crowd counting methods have been effective, head-scale variation and complex background remain two major challenges for crowd counting. Therefore, we propose a multiscale region calibration network called MRCNet to effectively address these challenges.

View Article and Find Full Text PDF

Epithelial cell collectives migrate through tissue interfaces and crevices to orchestrate development processes, tumor invasion, and wound healing. Naturally, the traversal of cell collective through confining environments involves crowding due to narrowing spaces, which seems tenuous given the conventional inverse relationship between cell density and migration. However, the physical transitions required to overcome such epithelial densification for migration across confinements remain unclear.

View Article and Find Full Text PDF

In order to solve the problem of weak single domain generalization ability in existing crowd counting methods, this study proposes a new crowd counting framework called Multi-scale Attention and Hierarchy level Enhancement (MAHE). Firstly, the model can focus on both the detailed features and the macro information of structural position changes through the fusion of channel attention and spatial attention. Secondly, the addition of multi-head attention feature module facilitates the model's capacity to effectively capture complex dependency relationships between sequence elements.

View Article and Find Full Text PDF

CMFX: Cross-modal fusion network for RGB-X crowd counting.

Neural Netw

December 2024

College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, China. Electronic address:

Currently, for obtaining more accurate counts, existing methods primarily utilize RGB images combined with features of complementary modality (X-modality) for counting. However, designing a model that can adapt to various sensors is still an unsolved issue due to the differences in features between different modalities. Therefore, this paper proposes a unified fusion framework called CMFX for RGB-X crowd counting.

View Article and Find Full Text PDF

Myelodysplastic syndromes (MDS) represent a group of bone marrow disorders involving cytopenias, hypercellular bone marrow, and dysplastic hematopoietic progenitors. MDS remains a challenge to treat due to the complex interplay between disease-induced and treatment-related cytopenias. Venetoclax, a selective BCL-2 inhibitor, in combination with azacitidine, a hypomethylating agent, is currently being investigated in patients with previously untreated higher-risk MDS.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!