Abstract: Inverse optimal control, also known as inverse reinforcement learning, is the problem o... more Abstract: Inverse optimal control, also known as inverse reinforcement learning, is the problem of recovering an unknown reward function in a Markov decision process from expert demonstrations of the optimal policy. We introduce a probabilistic inverse optimal control algorithm that scales gracefully with task dimensionality, and is suitable for large, continuous domains where even computing a full policy is impractical.
Abstract Interactive, task-guided character controllers must be agile and responsive to user inpu... more Abstract Interactive, task-guided character controllers must be agile and responsive to user input, while retaining the flexibility to be readily authored and modified by the designer. Central to a method's ease of use is its capacity to synthesize character motion for novel situations without requiring excessive data or programming effort. In this work, we present a technique that animates characters performing user-specified tasks by using a probabilistic motion model, which is trained on a small number of artist-provided animation clips.
Abstract This report presents a framework for studying the relationship between speech prosody an... more Abstract This report presents a framework for studying the relationship between speech prosody and motion of people in conversation. Several probabilistic models for the relationship between prosodic features in speech and the dynamics of body language are evaluated. These models are then applied to the task of synthesizing new body language animations for a novel utterance.
Abstract Human communication involves not only speech, but also a wide variety of gestures and bo... more Abstract Human communication involves not only speech, but also a wide variety of gestures and body motions. Interactions in virtual environments often lack this multi-modal aspect of communication. This thesis presents a method for automatically synthesizing body language animations directly from the participants' speech signals, without the need for additional input.
Abstract Artist-created animated characters can exhibit stylized, engaging behavior, but require ... more Abstract Artist-created animated characters can exhibit stylized, engaging behavior, but require considerable effort to construct, while interactive applications require numerous motions and variations to create a dynamic, believable character. This paper describes a method for generating some of these variations automatically: given a stream of poses, our method simulates plausible responses to physical disturbances and environmental variations.
Abstract: Inverse optimal control, also known as inverse reinforcement learning, is the problem o... more Abstract: Inverse optimal control, also known as inverse reinforcement learning, is the problem of recovering an unknown reward function in a Markov decision process from expert demonstrations of the optimal policy. We introduce a probabilistic inverse optimal control algorithm that scales gracefully with task dimensionality, and is suitable for large, continuous domains where even computing a full policy is impractical.
Abstract Interactive, task-guided character controllers must be agile and responsive to user inpu... more Abstract Interactive, task-guided character controllers must be agile and responsive to user input, while retaining the flexibility to be readily authored and modified by the designer. Central to a method's ease of use is its capacity to synthesize character motion for novel situations without requiring excessive data or programming effort. In this work, we present a technique that animates characters performing user-specified tasks by using a probabilistic motion model, which is trained on a small number of artist-provided animation clips.
Abstract This report presents a framework for studying the relationship between speech prosody an... more Abstract This report presents a framework for studying the relationship between speech prosody and motion of people in conversation. Several probabilistic models for the relationship between prosodic features in speech and the dynamics of body language are evaluated. These models are then applied to the task of synthesizing new body language animations for a novel utterance.
Abstract Human communication involves not only speech, but also a wide variety of gestures and bo... more Abstract Human communication involves not only speech, but also a wide variety of gestures and body motions. Interactions in virtual environments often lack this multi-modal aspect of communication. This thesis presents a method for automatically synthesizing body language animations directly from the participants' speech signals, without the need for additional input.
Abstract Artist-created animated characters can exhibit stylized, engaging behavior, but require ... more Abstract Artist-created animated characters can exhibit stylized, engaging behavior, but require considerable effort to construct, while interactive applications require numerous motions and variations to create a dynamic, believable character. This paper describes a method for generating some of these variations automatically: given a stream of poses, our method simulates plausible responses to physical disturbances and environmental variations.
Uploads