Recognizing Temporal Trajectories Using the Condensation Algorithm

The recognition of human gestures in image sequences isan importantand challengingproblem that enables a host of human-computerinteractionapplications. This paper describes an incremental recognition strategy that is an extension of the "Condensation" algorithm proposed by Isard and Blake (ECCV'96). Gestures are modeled as temporal trajectoriesof some estimated parameter over time (in this case velocity). The condensationalgorithmis used to incrementally match the gesture models to the inputdata. The methodis demonstrated with an example of an augmented office whiteboard in which a user makes simple hand gestures to grab regions of the board, print them, save them, etc.

• Type: Conference
• Year: 1998
• Where: FGR
• Authors: Michael J. Black, Allan D. Jepson
• Keywords: Biometrics, FGR 1998, Gestures, Human Gestures, Incremental Recognition Strategy