Mihai Polceanu

On Constrained Local Model Feature Normalization for Facial Expression Recognition

Real time user independent facial expression recognition is important for virtual agents but challenging. However, since in real time recognition users are not necessarily presenting all the emotions, some proposed methods are not applicable. In this paper, we present a new approach that instead of using the traditional base face normalization on whole face shapes, performs normalization on the point cloud of each landmark. The result shows that our method outperforms the other two when the user input does not contain all six universal emotions.

Computational mental simulation: A review

This paper is dedicated to the study of existing approaches that explicitly use mental simulation. Current implementations of the mental simulation paradigm, taken together, computationally address many aspects suggested by cognitive science research. Agents are able to find solutions to nontrivial scenarios in virtual or physical environments. Existing systems also learn new behavior by imitation of others similar to them and model the behavior of different others with the help of specialized models, culminating with the collaboration between agents and humans. Approaches that use self models are able to mentally simulate interaction and to learn about their own physical properties. Multiple mental simulations are used to find solutions to tasks, for truth maintenance, and contradiction detection. However, individual approaches do not cover all of the contexts of mental simulation and most rely on techniques which are only suitable for subsets of obtainable functionality. This review spans through four perspectives on the functionality of state‐of‐the‐art artificial intelligence applications, while linking them to cognitive science research results. Finally, an overview identifies the main gaps in existing literature on computational mental simulation and provides our suggestions for future development.

Simulation within simulation for agent decision-making

Generic computational model of mental simulation.The architecture proposes three modes - reactivity, predictability, and adaptability.First illustration shows a dog making predictions and testing different strategies.Second illustration is a 3D juggler plays with virtual jugglers, humans and robots. This article deals with artificial intelligence models inspired from cognitive science. The scope of this paper is the simulation of the decision-making process for virtual entities. The theoretical framework consists of concepts from the use of internal behavioral simulation for human decision-making. Inspired from such cognitive concepts, the contribution consists in a computational framework that enables a virtual entity to possess an autonomous world of simulation within the simulation. It can simulate itself (using its own model of behavior) and simulate its environment (using its representation of other entities). The entity has the ability to anticipate using internal simulations, in complex environments where it would be extremely difficult to use formal proof methods. Comparing the prediction and the original simulation, its predictive models are improved through a learning process. Illustrations of this model are provided through two implementations. First illustration is an example showing a shepherd, his herd and dogs. The dog simulates the sheeps behavior in order to make predictions testing different strategies. Second, an artificial 3D juggler plays in interaction with virtual jugglers, humans and robots. For this application, the juggler predicts the behavior of balls in the air and uses prediction to coordinate its behavior in order to juggle.

Regression and Mental Models for Decision Making on Robotic Biped Goalkeepers

We investigate the decision-making and behavior of robotic biped goalkeepers, applied to the RoboCup 3D Soccer Simulation League. We introduce two approaches to the goalkeepers behavior: first a heuristics-based approach that uses linear regression and Kalman filters for improved perception, and another based on mental models which uses nonlinear regression for ball trajectory filtering. Our experiments consist of 30,000 kick-and-save tests, using 100 random angle and distance kicks from six distance categories and four angle categories repeated 30 times. Our benchmark results show that both proposed approaches bring significant improvements for the goalkeepers save success rates

Towards A Theory-Of-Mind-Inspired Generic Decision-Making Framework

>

Real Time Tennis Match Tracking with Low Cost Equipment.

200i?ź% and validate the applicability of the novel mental model based decision-making process.

Soybean Plant Disease Identification Using Convolutional Neural Network.

We present in this work GRASPhere, a prototype device that enables users to select and manipulate on-screen objects with grasping gestures. We demonstrate GRASPhere as an extension of a widely-employed force-sensing device, the Phantom OMNI. The hardware design of GRASPhere employs a minimum of components that cost less than

An Interactive Virtual Training (IVT) Simulation for Early Career Teachers to Practice in 3D Classrooms with Student Avatars.

20, which makes our prototype easily replicable for practitioners interested in incorporating grasping gestures into their own force-feedback interactive applications. We discuss application opportunities for GRASPhere, such as exploring multimedia data with physical metaphors and providing assistance to people with visual impairments during indirect touch interaction.