Tracking subjects and detecting relationships in crowded city
videos

ELIÁŠ, Petr, Matúš MACKO, Jan SEDMIDUBSKÝ and Pavel ZEZULA. Tracking subjects and detecting relationships in crowded city videos. Multimedia Tools and Applications. Springer, 2022, Neuveden, January, p. 1-23. ISSN 1380-7501. Available from: https://dx.doi.org/10.1007/s11042-021-11891-z.

Other formats: BibTeX LaTeX RIS

TY  - JOUR
ID  - 1824299
AU  - Eliáš, Petr - Macko, Matúš - Sedmidubský, Jan - Zezula, Pavel
PY  - 2022
TI  - Tracking subjects and detecting relationships in crowded city videos
JF  - Multimedia Tools and Applications
VL  - Neuveden
IS  - January
SP  - 1-23
EP  - 1-23
PB  - Springer
SN  - 13807501
KW  - Multi-subject tracking;Relationship detection;2D skeleton sequences;Video analysis;Smart cities
UR  - https://link.springer.com/article/10.1007/s11042-021-11891-z
N2  - Multi-subject tracking in crowded videos is an established yet challenging research direction in computer vision and information processing. High applicability of multi-subject tracking is demonstrated in smart cities (e.g., public safety, crowd management, urban planning), autonomous driving vehicles, robotic vision, or psychology (e.g., social interaction and crowd behavior understanding). In this work, we propose a real-time approach that reveals tracks of subjects in ordinary videos, captured in highly populated pedestrian areas, such as squares, malls, and stations. The tracks are discovered based on the proximity of detected bounding boxes of subjects in consecutive video frames. The reduction of track fragmentation and identity switching is achieved by the re-identification phase that uses caching of unassociated detections and mutual projection of interrupted tracks. As the proposed approach does not require time-consuming extraction of appearance-based features, the superior tracking speed is achieved. In addition, we demonstrate tracker usability and applicability by extracting valuable information about body-joint positions from discovered tracks, which opens promising possibilities for detecting human relationships and interactions. We demonstrate accurate detection of couples based on their holding hand activity and families based on children's body proportions. The discovery of these entitative groups is especially challenging in crowded city scenes where many subjects appear in each frame.
ER  -

Basic information
Original name	Tracking subjects and detecting relationships in crowded city videos
Authors	ELIÁŠ, Petr (203 Czech Republic, guarantor, belonging to the institution), Matúš MACKO (703 Slovakia, belonging to the institution), Jan SEDMIDUBSKÝ (203 Czech Republic, belonging to the institution) and Pavel ZEZULA (203 Czech Republic, belonging to the institution).
Edition	Multimedia Tools and Applications, Springer, 2022, 1380-7501.

Other information
Original language	English
Type of outcome	Article in a journal
Field of Study	10200 1.2 Computer and information sciences
Country of publisher	United States of America
Confidentiality degree	is not subject to a state or trade secret
WWW	URL
Impact factor	Impact factor: 3.600
RIV identification code	RIV/00216224:14330/22:00129021
Organization unit	Faculty of Informatics
Doi	http://dx.doi.org/10.1007/s11042-021-11891-z
UT WoS	000743891900007
Keywords in English	Multi-subject tracking;Relationship detection;2D skeleton sequences;Video analysis;Smart cities
Tags	DISA
Tags	International impact, Reviewed
Changed by	Changed by: doc. RNDr. Jan Sedmidubský, Ph.D., učo 60474. Changed: 10/3/2023 12:58.

Abstract

Multi-subject tracking in crowded videos is an established yet challenging research direction in computer vision and information processing. High applicability of multi-subject tracking is demonstrated in smart cities (e.g., public safety, crowd management, urban planning), autonomous driving vehicles, robotic vision, or psychology (e.g., social interaction and crowd behavior understanding). In this work, we propose a real-time approach that reveals tracks of subjects in ordinary videos, captured in highly populated pedestrian areas, such as squares, malls, and stations. The tracks are discovered based on the proximity of detected bounding boxes of subjects in consecutive video frames. The reduction of track fragmentation and identity switching is achieved by the re-identification phase that uses caching of unassociated detections and mutual projection of interrupted tracks. As the proposed approach does not require time-consuming extraction of appearance-based features, the superior tracking speed is achieved. In addition, we demonstrate tracker usability and applicability by extracting valuable information about body-joint positions from discovered tracks, which opens promising possibilities for detecting human relationships and interactions. We demonstrate accurate detection of couples based on their holding hand activity and families based on children's body proportions. The discovery of these entitative groups is especially challenging in crowded city scenes where many subjects appear in each frame.

Links
GA19-02033S, research and development project	Name: Vyhledávání, analytika a anotace datových toků lidských pohybů
GA19-02033S, research and development project	Investor: Czech Science Foundation

PrintDisplayed: 27/4/2024 11:56

Tracking subjects and detecting relationships in crowded city videos

Other applications