S. Intille and A. Bobick. Representation and visual recognition of complex multi-agent actions using Belief networks. In ECCV Workshop on Perception of Human Action, Freiburg, Germany, June 1998. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
On the Semantics of Visual Behaviour, Structured Events and.. - Gong, Ng, Sherrah (Correct)
....Furthermore, changes in the structure of a representation alters the underlying context therefore its semantics. This can be modelled as belief revision [16] Com putationally, Bayesian belief networks have been widely adopted for the task of encoding knowledge as semantics of visual behaviour [20,25,39,40,47,48]. Alternatively, Ivanov and Bobick [28] proposed to use stochastic grammar to describe high level behaviour. Their approach tried to learn grammars from data rather than specifying them manually. What they did have to specify was the so called atomic semantic units . We consider these atomic ....
S. Intille and A. Bobick. Representation and visual recognition of complex multi-agent actions using Belief networks. In ECCV Workshop on Perception of Human Action, Freiburg, Germany, June 1998.
General Examination on Technical Area: Recognizing human activity .. - Sawhney (Correct)
.... towards visual sensor fusion, combining the output of several vision algorithms in an interactive kiosk to detect the presence of a speaker [Rehg99] They have also been applied towards visual surveillance and analysis tasks, requiring the recognition of complex multi agent actions in a scene [Intille98]. Figure 4: A Dynamic Bayesian network (DBN) showing control layer states X 1 . X 3 factored into composite variable nodes S 1 ,S 2 chained together at each time step. Factored States (sub model layer) X 1 S 1 X 2 X 3 S 2 S 1 S 2 S 2 Control layer Observations S 1 Y 1 Y 2 Y 3 7 3. ....
Intille, S.S. and Bobick A. 1998. Representation and visual recognition of complex, multiagent actions using belief networks. IN CVPR '98 Workshop on Interpretations of Visual Motion. Also see MIT Media Lab TR 454.
Multimodal Speaker Detection using Input/Output Dynamic.. - Pavlovic, Garg, Rehg (Correct)
....combine an intuitive graphical representation with efficient algorithms for inference and learning. Previous work has demonstrated the power of these models in fusing video and audio cues with contextual information and expert knowledge both for speaker detection and other similar applications [3, 5, 4, 2]. Speaker detection is a particularly interesting example of a multi modal sensing task with application in video conferencing, video indexing and human computer interaction. Both video and audio sensing provide important information in a multi person and noisy scenarios. Contextual information ....
S. Intille and A. Bobick, "Representation and visual recognition of complex, multi-agent actions using belief networks," Tech. Rep. 454, MIT Media Lab, Cambridge, MA, 1998.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC