Detecting Faces in Images: A Survey
IEEE Transactions on Pattern Analysis and Machine Intelligence
Probabilistic Methods for Finding People
International Journal of Computer Vision
Robust Face Tracking Using Color
FG '00 Proceedings of the Fourth IEEE International Conference on Automatic Face and Gesture Recognition 2000
HUMO '00 Proceedings of the Workshop on Human Motion (HUMO'00)
Human Detection Using Depth and Gray Images
AVSS '03 Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance
Grouped-People Splitting Based on Face Detection and Body Proportion Constraints
AVSS '06 Proceedings of the IEEE International Conference on Video and Signal Based Surveillance
Implementation of real-time video conference system using high speed multimedia communication
WSEAS Transactions on Computers
| Hi-index | 0.00 |
Counting people in a video stream for applications such as video surveillance or imaging statistical systems is considered a challenging task since video streams require complex computations and affected by noisy environment. This paper proposes a hybrid model for counting people using a video stream based on human skin detection and geometrical head recognition. This model is based on detecting single and multiple head-counts, while in motion by positioning the camera in different angle setups. A 45 and 90 camera positions will detect both the skin color information and the head shape of an object simultaneously. The experiments results indicated that the model accurately detects and counts people on real world environments and is robust to changes in viewpoint, scale, and object speed.