Face Transfer is a method for mapping videorecorded performances of one individual to facial animations of another. It extracts visemes (speech-related mouth articulations), expressions, and three-dimensional (3D) pose from monocular video or film footage. These parameters are then used to generate and drive a detailed 3D textured face mesh for a target identity, which can be seamlessly rendered back into target footage. The underlying face model automatically adjusts for how the target performs facial expressions and visemes. The performance data can be easily edited to change the visemes, expressions, pose, or even the identity of the target—the attributes are separably controllable. This supports

Abstract — Nonverbal communication plays an important role in coordinating teammates ’ actions for collaborative activities. In this paper, we explore the impact of non-verbal social cues and behavior on task performance by a human-robot team. We report our results from an experiment where naïve human subjects guide a robot to perform a physical task using speech and gesture. The robot communicates either implicitly through behavior or explicitly through non-verbal social cues. Both selfreport via questionnaire and behavioral analysis of video offer evidence to support our hypothesis that implicit non-verbal communication positively impacts human-robot task performance with respect to understandability of the robot, efficiency of task performance, and robustness to errors that arise from miscommunication. Whereas it is already well accepted that social cues enhance the likeability of robots and animated agents, our results offer promising evidence that they can also serve a pragmatic role in improving the effectiveness human-robot teamwork where the robot serves as a cooperative partner.

We contribute an approach for interactive policy learning through expert demonstration that allows an agent to actively request and effectively represent demonstration examples. In order to address the inherent uncertainty of human demonstration, we represent the policy as a set of Gaussian mixture models (GMMs), where each model, with multiple Gaussian components, corresponds to a single action. Incrementally received demonstration examples are used as training data for the GMM set. We then introduce our confident execution approach, which focuses learning on relevant parts of the domain by enabling the agent to identify the need for and request demonstrations for specific parts of the state space. The agent selects between demonstration and autonomous execution based on statistical analysis of the uncertainty of the learned Gaussian mixture set. As it achieves proficiency at its task and gains confidence in its actions, the agent operates with increasing autonomy, eliminating the need for unnecessary demonstrations of already acquired behavior, and reducing both the training time and the demonstration workload of the expert. We validate our approach with experiments in simulated and real robot domains.

Abstract not found

Head pose and gesture offer several key conversational grounding cues and are used extensively in face-to-face interaction among people. We investigate how dialog context from an embodied conversational agent (ECA) can improve visual recognition of user gestures. We present a recogntion framework which (1) extracts contextual features from an ECA´s dialog manager, (2) computes a predicition of head nod and head shakes, and (3) integrates the contextual predictions with the visual observation of a vision-based head gesture recognizer. We found a subset of lexical, punctuation and timing features that are easily available in most ECA architectures and can be used to learn how to predict user feedback. Using a discriminative approach to contextual prediction and multi-modal integration, we were able to improve the performancae of head gesture detection even when the topic of the test set was significantly different than the training set.

Abstract — In their interactions with the world, robots inevitably face situations in which multiple actions are equivalently applicable. These situations violate the common assumption that for any world state there exists a single best action. When learning from demonstration, this ambiguity frequently results in inconsistent demonstrations from the teacher, however, the problem of action choices has been overlooked by previous approaches for demonstration learning. In this paper, we present an algorithm that identifies regions of the state space with conflicting demonstrations and enables the choice between multiple actions to be represented explicitly within the robot’s policy. An experimental evaluation of the algorithm in a real-world obstacle avoidance domain shows that reasoning about action choices significantly improves the robot’s learning performance. I.

Recognizing and responding to human affect is important in collaborative tasks in joint human-robot teams. In this paper we present an integrated architecture for HRI and report results from an experiment with this architecture that shows that expressing affect and responding to human affect with affect expressions improves performance in a joint human-robot task. 1.

The use of the computer as a model, metaphor, and modelling tool has tended to privilege the ‘cognitive ’ over the ‘affective ’ by engendering theories in which thinking and learning are viewed as information processing and affect is ignored or marginalised. In the last decade there has been an accelerated flow of findings in multiple disciplines supporting a view of affect as complexly intertwined with cognition in guiding rational behaviour, memory retrieval, decision-making, creativity, and more. It is time to redress the imbalance by developing theories and technologies in which affect and cognition are appropriately integrated with one another. This paper describes work in that direction at the MIT Media Lab and projects a large perspective of new research in which computer technology is used to redress the imbalance that was caused (or, at least, accentuated) by the computer itself. 1. Vision The last half-century of technological acceleration has yielded a massive incursion of digital technology into the learning environment, making dramatic differences, and promising even greater changes, to the practice of learning. Computers have served as tools to aid in learning at all levels from simple classroom activities to the way theorists think about thinking. The field of artificial intelligence, with emphasis on ideas such

Natural human-like human-robot interaction (NHL-HRI) requires the robot to be skilled both at recognizing and producing many subtle human behaviors, often taken for granted by humans. We suggest a rough division of these requirements for NHL-HRI into three classes of properties: (1) social behaviors, (2) goal-oriented cognition, and (3) robust intelligence, and present the novel DIARC architecture for complex affective robots for human-robot interaction, which aims to meet some of those requirements. We briefly describe the functional properties of DIARC and its implementation in our ADE system. Then we report results from human subject evaluations in the laboratory as well as our experiences with the robot

To make the transition from a technological curiosity to productive tools, humanoid robots will require key advances in many areas, including, mechanical design, sensing, embedded avionics, power, and navigation. Using the NASA Johnson Space Center’s Robonaut as a testbed, the DARPA Mobile Autonomous Robot Software (MARS) Humanoids team is investigating technologies that will enable humanoid robots to work effectively with humans and autonomously work with tools. A novel learning approach is being applied that enables the robot to learn both from a remote human teleoperating the robot and an adjacent human giving instruction. When the remote human performs tasks teleoperatively, the robot learns the salient sensory-motor features for executing the task. Once learned, the task may be carried out by fusing the skills required to perform the task, guided by on-board sensing. The adjacent human takes advantage of previously learned skills to sequence the execution of these skills. Preliminary results from initial experiments using a drill to tighten lug nuts on a wheel are discussed.