An Effective Pipeline for Pedestrian Detection in Mid-High Density Crowd
- Submitting institution
-
The University of Huddersfield
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 10
- Type
- E - Conference contribution
- DOI
-
10.23919/IConAC.2019.8894985
- Title of conference / published proceedings
- 25th International Conference on Automation & Computing (ICAC 2019) : Improving Productivity through Automation and Computing
- First page
- 153
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- November
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Sponsored by the Chinese National Natural Science Foundation (No. 61671377) and Shaanxi Province (2019GY-054), this paper's innovative spatio-temporal feature descriptor for video-based medium-to-high-density crowd analysis was developed in partnership with staff from Zhejiang University (World Ranked no.53 in 2021 https://www.topuniversities.com/universities/zhejiang-university) and ShaanXi Electronic Information Scene Investigation Centre. The research enables complex crowd scenarios to be classified through recognition of changes against context and surroundings. The classifier's high-fidelity global motion pattern descriptions have been adopted in live systems in 8 Chinese cities (https://www.sohu.com/a/421649772_120159460) and received a prestigious scientific break-through prize in 2020 (https://helios.hud.ac.uk/sengdg2/2-provincial-award-certificate.pdf)
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -