Overview
The SUT Anomaly Detection Dataset was collected and created by Habib Ebadi Namin at Shahrood University of Technology to provide a high-resolution benchmark for single-scene video anomaly detection. The dataset comprises 86 videos containing a total of 26,483 frames, divided into 59 training videos (14,642 frames) and 27 testing videos (11,841 frames). All frames have a spatial resolution of 1920 × 1080 pixels.
The training set contains only normal activities and intentionally includes natural background appearance variations, such as changes in indoor illumination and scene decoration, to improve the robustness of anomaly detection models.
The testing set contains both normal and anomalous events and is accompanied by frame-level ground-truth annotations for quantitative evaluation. Running events occur both toward/away from and across the surveillance camera, producing significant object-scale variations due to perspective.
Dataset Statistics
| Property | Training | Testing | Total |
|---|---|---|---|
| Videos | 59 | 27 | 86 |
| Frames | 14,642 | 11,841 | 26,483 |
| Resolution | 1920 × 1080 | ||
| Format | PNG | MP4 (H.264) | — |
| Annotation | — | Frame-level | — |
Sample Frames
Citation & Acknowledgments
BibTeX template will populate here.
Special thanks to Dr. Alireza Ahmadifard for supervision, and to Dr. Reza Kharghanian for feedback and constructive advice.
Contact Information
- Habib Ebadi Namin (
HabibEbadi@ieee.org) - Dr. Alireza Ahmadifard (
ahmadyfard@shahroodut.ac.ir) - Dr. Reza Kharghanian (
reza.kharghanian@shahroodut.ac.ir)