The Hand Mouse: GMM hand-color classification and mean shift tracking

Vision-Based Hand Gesture Recognition for Human-Computer Interaction

“…For increased robustness, the method tracks the centroid of the blob and also continuously adapts the representation of the tracked color distribution. Similar is also the method proposed in [KOKS01], except the fact that it utilizes a Gaussian mixture model to approximate the color histogram and the EM algorithm to classify skin pixels based on the Bayesian decision theory.…”

“…A simple color comparison scheme is employed in [DKS01], where the dominant color of a homogeneous region is tested as if occurring within a color range that corresponds to skin color variability. Other approaches [Bra98,KOKS01,MC97] consider skin color to be uniform across image space and extract the pursued regions through typical region-growing and pixel-grouping techniques. More advanced color segmentation techniques rely on histogram matching [Ahm94], or employ a simple look-up table approach [KK96,QMZ95] based on the training data for the skin and possibly its surrounding areas.…”

An Introduction to the 3rd Workshop on Egocentric (First-Person) Vision

“…Much of the early work from the late 90s using computer vision techiques were focused on the wearable camera as a gestural interface [40,21,42,69,70]. While early work helped highlight the opportunity to develop computer vision algorithms for mobile egocentric vision systems, the focus was primarily on analysis of fingers, hands and faces [70,29,37,40].…”

Handy AR: Markerless Inspection of Augmented Reality Objects Using Fingertip Tracking

“…Many early hand-based UIs rely on tracking the hand in a 2D viewing plane and use this information as a mouse replacement [17] [18]. Some systems demonstrated fingertip tracking for interactions in a desktop environment [20] [21].…”