International Journal of Advanced Innovative Technology in Engineering (IJAITE)

---

State of the Art Human Activity Detection

Veena K. Katankar, Piyush A. Fukat, Swapnil S. Urade, Tushar S. Tajne, Puja S. Delikar, Arti M. Waghmare

Abstract :

The mission of the Human Fight Detection is to control the growing violence in the day-to-day life of humans by implementing the appropriate violence detecting algorithms. Surveillance scenarios like prisons, psychiatric centers, or even embedded in camera phones, but cannot be more capable to solve the problem or cannot be more effective to take action on needed time such types of violence are now growing tremendously. As a consequence, there is rising interest in developing violence detection algorithms. Recent work is considered by the well-known Bag of Words framework for the specific problem of violence detection. Under the framework of this, spatial-temporal features are extracted from the video sequences and used for classification. The dataset of violence collected consists of fight scenes from surveillance camera videos available in sources online platforms. The dataset is made publicly available. From the extensive experiments conducted on Hockey Fight, Pellicle, and the newly collected fights datasets, the overall research has been totally helpful for us in finding the new potential for future references.

Keywords :

Full Text :

Download PDF

DOI :

Cite this paper :

References :

---