Experiments of Crowd Detection for Crowd Digital Twins
The development of a crowd digital twin offers significant potential for enhancing public safety, urban planning, and event management. A key challenge in creating such a digital twin lies in the efficient and accurate acquisition of crowd-related data, particularly through object detection models deployed on resource-constrained devices. Through a series of experiments, we compare TinyML and Edge approaches in terms of detection accuracy, inferencing time, and resource utilization. Our findings highlight the trade-offs inherent in selecting detection models for crowd digital twin applications, underscoring the importance of aligning model choice with specific deployment needs.
Authors
- Yue Yu (ORCID: https://orcid.org/0000-0003-3529-585X)
- A. C. Testa (ORCID: https://orcid.org/0000-0003-2217-8726)
- Silvia Mirri (ORCID: https://orcid.org/0000-0002-5385-4734)
- Roberto Girau (ORCID: https://orcid.org/0000-0002-1298-7960)
- Giovanni Delnevo (ORCID: https://orcid.org/0000-0001-6640-5746)
- Su-Kit Tang (ORCID: https://orcid.org/0000-0001-8104-7887)
- Rui Wang (ORCID: https://orcid.org/0009-0002-5248-9955)
- Roberto Casadei (ORCID: https://orcid.org/0000-0001-9149-949X)
- Chan‐Tong Lam (ORCID: https://orcid.org/0000-0002-8022-7744)
- Rui Wang (ORCID: https://orcid.org/0009-0008-8726-0601)
- Connie Lai (ORCID: https://orcid.org/0000-0002-8951-6788)
- Kuan Pok Chong
- Weibo Ling
- Zhuoqian Lu
Institutions
- Guangdong University of Foreign Studies (CN)
- Macao Polytechnic University (MO)
- University of Bologna (IT)
Publication Details
- Published
- 2025-01-10
- DOI
- https://doi.org/10.1109/ccnc54725.2025.10976164
- Primary Topic
- Mobile Crowdsensing and Crowdsourcing
- Type
- article
- Field-Weighted Citation Impact
- 0.00