Multi-Camera Multi-Person Pointing Gesture Recognition for Interaction in Immersive Environments


Arcot Sowmya.

Student: Anuraag Sridhar.
Introduction: Immersive environments is a field of research that has greatly benefited from recent advances in hardware and software technologies.
Significance: They have proven to be quite useful in fields ranging from entertainment to education. As such environments improve, there is also a constant need to improve the interaction paradigm within the immersive application. This problem has been tackled in many different ways from virtual reality gear through to computer vision based interaction.
Applications: In this research we demonstrate a system which uses multiple cameras to allow multiple people to interact with a large-scale immersive display, via a pointing gesture. The cameras are separated into two basic subsets - overhead cameras and side cameras. Overhead cameras are used to perform multi-person tracking, and side cameras are used to perform posture recognition.
Challenges: Vision-based interaction provides the least encumbered mechanism for interaction as it removes the reliance on wires, markers and wearable outfits.
top of page