Interessante que o Jacob tem pesquisa vinculada.

MPX is a way to have multiple pointers in a X-Windows system. The usual X-Windows systems supports multiple input devices, but just one pointer.

What about FocusShare in Windows? Does it already the distribution we need? And can we combine that with CPNmouse which supports multiple mice in Windows (see also the sources here)?

• The Hand Mouse: GMM hand-color classification and mean shifttracking (2001)
• VisualPanel: Toward A Vision-Based Mobile Input Interface For Anywhere (2000)

Characteristics: One camera, panel detection by contrast (vague), dynamic programming quadrangle tracking, background difference for top detection (vague), conic fitting of tip pointer.

• Fingermouse: A wearable hand tracking system (2002)
• DiamondTouch: A Multi-User Touch Technology (2003)
• Multi-finger Cursor Techniques (2006)

Characteristics: Input is standard touchpad. Describes new cursor techniques for >1 finger.

Approaches: #cameras (2D,3D), type of camera (vision,infrared,…), type of object (hand, any, marked with color, etc.), type of model(shape, skin-color, etc., if any)

• Visual tracking of high DOF articulated structures: an application to human hand tracking (1994)
• A usable real-time 3D hand tracker (1994)
• Real-time hand tracking and gesture recognition using smart snakes (1995)
• DigitEyes: vision-based hand tracking for human-computerinteraction (1999)
• Fast tracking of hands and fingertips in infrared images for augmented desk interface (2000)
• Hand tracking for human-computer interaction with Graylevel VisualGlove: turning back to the simple way (2001)
• Real-Time Fingertip Tracking and Gesture Recognition (2002)
• SmartSkin - An Infrastructure for Freehand Manipulation on Interactive Surfaces (2002)
• The perceptive workbench: Computer-vision-based gesture tracking, object tracking, and 3D (2003)
• Real-time Hand Tracking and Finger Tracking for Interaction
• Real time hand tracking by combining particle filtering and mean shift (2004)
• Fast 2D Hand Tracking with Flocks of Features and Multi-Cue Integration (2004)

Characteristics: One camera. Pyramid-based KLT extract ~100 features (sharp bi-gradients), then flock model (birds) tracks their motion; color for features is learned.

• Visual Touchpad: A Two-handed Gestural Input Device (2004)

Characteristics: Two cameras. Based on VisualPanel.

• Automatic 2D Hand Tracking in Video Sequences (2005)

Characteristics: Tracks hands (finds bounding box of two hands in videos).

• Interacting with large displays from a distance with vision-tracked multi-finger gestural input (2005)

Characteristics: Two-handed, two cameras, black input area (with id tag), gestures, based on Visual Touchpad.

Well, there is no lack of publications in this field…