JEBARA,T.,AND PENTLAND, A. 1999. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In International Conference on Computer Vision Systems (ICVS), vol. 1542, 273--292.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....[131] Most of their work has focused on the low level sensing infrastructure and HCI issues [38, 114, 110] although [102] uses stochastic grammars to parse videos. The MIT Media Lab hosts a number of interrelated ubiquitous computing efforts, including Smart Rooms and Learning Humans [117, 118, 14, 78]. Their work, like ours, is based on probabilistic behavior models, but to date they have not considered hierarchies of plans and goals or user errors. Media lab projects specific to problems of the elderly include wearable medical monitors [129] and Memory Glasses [37] Some of these systems will ....

Tony Jebara and Alex Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In Proceedings of the International Conference on Vision Systems (ICVS'99), Las Palmas de Gran Canaria, Spain, 1999.

....by Oliver, Rosario and Pentland [34] 35] has developed reliable Bayesian vision systems. Two exciting recent developments are: 1) work by Galata, Johnson and Hogg using deformable models with HMM behaviour models for virtual actors [24] and 2) the action reaction learning of Jebara and Pentland [36], 37] which models interactions and exploits new ideas from Support Vector Machines (SVMs) in conjunction with generative Bayesian theory. D.1 Visual representation The main system we look at in some detail here, as an exemplar of graphical models, is the HIVIS watcher system [38] 39] When ....

T. Jebara and A. Pentland, \Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour," in International conference on Vision Systems, Gran Canaria, Spain, 1999, pp. 273-292.

....2. 1 Overview There is growing interest in computer vision and multimedia signal processing for understanding the behaviour of interacting people, for actions that are de ned by playing both similar and complementary roles (e.g. a handshake, a dancing couple, or a children s game) 5] [6], 7] 8] While most of the work for recognition of interactions has been directed towards visual surveillance in outdoor [7] and oce scenarios [6] the analysis of people interaction constitutes a richer research domain. Group interaction recognition can be approached probabilistically with ....

.... people, for actions that are de ned by playing both similar and complementary roles (e.g. a handshake, a dancing couple, or a children s game) 5] 6] 7] 8] While most of the work for recognition of interactions has been directed towards visual surveillance in outdoor [7] and oce scenarios [6], the analysis of people interaction constitutes a richer research domain. Group interaction recognition can be approached probabilistically with models that handle multiple information streams and capture consistent data relationships. Within this framework, interaction recognition can be ....

T. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In Proc. International Conference on Vision Systems, January 1999.

....provide real time human body analy sis. Gesture recognition then involves techniques such as Hidden Markov models (HMMs) which can be parameterised to provide information such as direction of pointing [21] or human interaction analysis using the Conditional Expectation Maximisation (CEM) approach [14]. Recently, we have been developing computationally simple view based approaches to action recognition under the ISCANIT project. These new systems start to address the task of intentional tracking and behavioural modelling to drive visual interaction. Our original work centred around ....

A. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In Proceedings of International Conference on Vision Systems (ICV$'99), Las Palmas de Gran Canaria, Spain, 1999.

.... device for mediating interaction between humans and machines, enabling a form of interaction that mimics direct communication between humans [10, 1, 11] The experience of realism is further enhanced when the computer is equipped with visual and auditory senses with which to perceive the user [6, 2, 13, 4]. In this symmetric situation, both the human and synthetic head can see and be seen, and can hear and be heard. Of paramount importance within face to face conversation is of course the content of what is said. However, the sequence and timing of accompanying facial expressions is also ....

....mistimed or inappropriate expressions may convey unintended meaning and can therefore be disruptive. It is reasonable to suppose the same requirements will apply for human interaction with a synthetic talking head. An approach that begins to meet these requirements is proposed in [9] and [6]. Their idea is based on the common notion of a state space, in which each vector represents the instantaneous configuration of a participant in an interaction. Such vectors are the endpoint of a perceptual process within the computer, sensing the human party, and the startpoint for a graphical ....

[Article contains additional citation context not shown here]

T. Jebara and A. Pentland. Action reaction learning : Automatic visual analysis and synthesis of interactive behaviour. In ICVS'99, pages 273--292, January 1999.

....last decade it has been extensively applied in a large number of problems. A non exhaustive list of such problems consists of speech recognition [3, 6] handwritten character recognition [7] DNA and protein modelling [8] gesture recognition [9] and, more recently, behavior analysis and synthesis [10]. The use of HMM for shape analysis has not been widely addressed. Only a few work have been found to have some similarities with our approach. In the first, He and Kundu [17] utilize HMMs to model shape contours represented by auto regressive (AR) coefficients. Results are quite interesting and ....

Jebara, T., Pentland, A.: Action Reaction Learning: Automatic Visual Analysis and Synthesis of interactive behavior. In 1st Intl. Conf. on Computer Vision Systems (ICVS'99) (1999).

....last decade it has been extensively applied in a large number of problems. A non exhaustive list of such problems consists of speech recognition [1] handwritten character recognition [4] DNA and protein modelling [5] gesture recognition [6] and, more in general, behavior analysis and synthesis [7]. HMMs t very well in a large number of situations, in particular where the state sequence structure of the process examined can be assumed to be Markovian. Unfortunately, there are some drawbacks [8] First, the iterative technique for the HMM learning (Baum Welch re estimation) converges to a ....

Jebara, T., Pentland, A.: Action Reaction Learning: Automatic Visual Analysis and Synthesis of interactive behavior. In 1st Intl. Conf. on Computer Vision Systems (ICVS'99) (1999).

....can be learned by repetitive training. Examples, in the realm of visual recognition, are the works of Starner and Pentland [165] Wilson and Bobick [181] and Brand et al. 27] Recently there has been some work with HMMs to represent human interaction [118] and human machine interaction [70]. Intille and Bobick [63] have employed a combination of a frame based language and Bayesian networks to represent and recognize American football plays. This is one of the few recognition methods that explicitly acknowledges the need for and investigates the issue of considering both context and ....

T. Jebara and A. Pentland. "Action Reaction Learning: Automatic Visual Analysis and Synthesis of Interactive Behavior", Proc. of the International Conference on Computer Vision Systems (ICVS'99), Las Palmas, Gran Canaria, Spain, pp. 273-292. January 1315. 1999.

No context found.

T. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In International Conference on Vision Systems, 1999.

No context found.

T. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In International Conference on Vision Systems, 1999.

....MAPPING Figure 1.5: The AIM mapping from action to perception. An alternative approach is to do away with the AIM problem altogether by either providing the teacher s perceptual data in terms of the action space of the learner [201] 124] or by only considering virtual characters [74] 61] 101] [93] whose action space is in the perceptual space. For example, Weng [201] describes a human pushing a robot down a hallway while the robot collects images of its context. The actuators in the robot (not its cameras) measure the human s displacement and therefore to imitate the human, the ....

T. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In International Conference on Vision Systems, 1999.

..... Learning systems have the ability to learn new behaviors or to modify the existing ones by dynamically modifying parameters of the original behaviors. These systems provide a rich set of interaction modalities and dynamics, and offer new interesting venues for interactive media architectures [9]. Perceptive systems are modeled according to a new way of thinking about and authoring interactive media which we present briefly in this section and expand in a later paragraph. We introduce an additional layer in the one to many mapping between sensory input and system response, called the ....

Jebara T., Pentland A., "Action Reaction Learning: Automatic Visual Analysis and Synthesis of Interactive Behaviour", ICVS '99, Las Palmas, Canary Islands, January 1999.

No context found.

JEBARA,T.,AND PENTLAND, A. 1999. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In International Conference on Computer Vision Systems (ICVS), vol. 1542, 273--292.

No context found.

T. Jebara and A. Pentland, "Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour," in Proc. International Conference on Vision Systems, January 1999.

No context found.

T. Jebara and A. Pentland, "Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour," in Proc. International Conference on Vision Systems, January 1999.

No context found.

T. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In Proc. 1st Int. Conf. on Computer Vision Systems, volume 1542 of Lecture Notes in Computer Science, pages 273--292, 1999.

No context found.

T. Jebara and A. Pentland, "Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour," in Proc. International Conference on Vision Systems, January 1999.

No context found.

T. Jebara and A. Pentland. Action reaction learning : Automatic visual analysis and synthesis of interactive behaviour. In ICVS'99, pages 273--292, January 1999.

No context found.

T. Jebara and A. Pentland, "Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour," in Proc. International Conference on Vision Systems, January 1999.

No context found.

T. Jebara and A. Pentland. Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour. In Proc. International Conference on Vision Systems, January 1999.

No context found.

Jebara, T., Pentland, A.: Action Reaction Learning: Automatic Visual Analysis and Synthesis of interactive behavior. In 1st Intl. Conf. on Computer Vision Systems (ICVS'99) (1999).

No context found.

T. Jebara and A. Pentland, \Action reaction learning: Automatic visual analysis and synthesis of interactive behaviour," in Proc. International Conference on Vision Systems, January 1999.

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