Hand Tracking and Estimation Demo video
We present a novel system which recognises sign language hand shapes from video data captured in 2D. By mapping the recognised hand shape from 2D to 3D, it is possible to obtain 3D co-ordinates of each of the joints within the hand using the kinematics embedded in a 3D avatar hand and smooth the transformation in 3D space between any given hand shapes. The novelty in our system is that it does not require a hand pose to be recognised at every frame, but rather that hand shapes be detected at a given step size. This architecture allows for a more efficient system with better accuracy than other related systems.
Number Plate Detection Demo video
Numberplate detection plays an important role in public transportation systems. Before numberplate detection can be performed, the region of interest that contains the numberplate needs to be determined. This paper presents a novel approach todetecting numberplates in a video sequence by multisegmentation. The first segmentation determines the approximate location of the numberplate by determining the distance between each of the letters and numbers found on the number plate. This feature is unique to numberplates and therefore provides an excellent starting point in detecting numberplates. The second segmentation focuses on the region of interest provided by the first segmentation. This includes edge detection and, a horizontal and vertical projection function that determines the boundary of the numberplate.