Conference paper Open Access
Spatio-temporal activity detection and recognition in untrimmed surveillance videos
Konstantinos Gkountakos;
Despoina Touska;
Konstantinos Ioannidis;
Theodora Tsikrika;
Stefanos Vrochidis;
Ioannis Kompatsiaris
This work presents a spatio-temporal activity detection and recognition framework for untrimmed surveillance videos consisting of a three-step pipeline: object detection, tracking, and activity recognition. The framework relies on the YOLO v4 architecture for object detection, Euclidean distance for tracking, while the activity recognizer uses a 3D Convolutional Deep learning architecture employing spatio-temporal boundaries and addressing it as multi-label classification. The evaluation experiments on the VIRAT dataset achieve accurate detections of the temporal boundaries and recognitions of activities in untrimmed videos, with better performance for the multi-label compared to the multi-class activity recognition.
| Name | Size | |
|---|---|---|
|
3.Spatio-temporal activity detection and recognition in untrimmed surveillance videos_Zenodo.pdf
md5:966ffebedd56853d75ef4e34a553e1b8 |
1.0 MB | Download |
| All versions | This version | |
|---|---|---|
| Views | 199 | 199 |
| Downloads | 155 | 155 |
| Data volume | 156.3 MB | 156.3 MB |
| Unique views | 179 | 179 |
| Unique downloads | 144 | 144 |
- Publication date:
- May 11, 2021
- DOI:
-
- Keyword(s):
- Activity detection Activity recognition Spatiotemporal boundaries detection 3D-convolutional neural networks
- Grants:
-
European Commission:
-
PREVISION - Prediction and Visual Intelligence for Security Information (833115)
-
CREST - Fighting Crime and TerroRism with an IoT-enabled Autonomous Platform based on an Ecosystem of Advanced IntelligEnce, Operations, and InveStigation Technologies (833464)
- CONNEXIONs - InterCONnected NEXt-Generation Immersive IoT Platform of Crime and Terrorism DetectiON, PredictiON, InvestigatiON, and PreventiON Services (786731)
-
PREVISION - Prediction and Visual Intelligence for Security Information (833115)
- Communities:
- License (for files):
- Creative Commons Attribution 4.0 International