Recognizing Facial Expressions Using Model-Based Image Interpretation
Multimodal Signals: Cognitive and Algorithmic Issues
A Model Based Approach for Expressions Invariant Face Recognition
ICB '09 Proceedings of the Third International Conference on Advances in Biometrics
Did I Get It Right: Head Gestures Analysis for Human-Machine Interactions
Proceedings of the 13th International Conference on Human-Computer Interaction. Part II: Novel Interaction Methods and Techniques
Refining the execution of abstract actions with learned action models
Journal of Artificial Intelligence Research
Generality and legibility in mobile manipulation
Autonomous Robots
Action-related place-based mobile manipulation
IROS'09 Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems
Adjusted pixel features for robust facial component classification
Image and Vision Computing
Facial expression recognition for human-robot interaction: a prototype
RobVis'08 Proceedings of the 2nd international conference on Robot vision
Face recognition using wireframe model across facial expressions
BioID_MultiComm'09 Proceedings of the 2009 joint COST 2101 and 2102 international conference on Biometric ID management and multimodal communication
ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
Learning and reasoning with action-related places for robust mobile manipulation
Journal of Artificial Intelligence Research
Image and Vision Computing
Hi-index | 0.14 |
Model-based techniques have proven to be successful in interpreting the large amount of information contained in images. Associated fitting algorithms search for the global optimum of an objective function, which should correspond to the best model fit in a given image. Although fitting algorithms have been the subject of intensive research and evaluation, the objective function is usually designed ad hoc, based on implicit and domain-dependent knowledge. In this article, we address the root of the problem by learning more robust objective functions. First, we formulate a set of desirable properties for objective functions and give a concrete example function that has these properties. Then, we propose a novel approach that learns an objective function from training data generated by manual image annotations and this ideal objective function. In this approach, critical decisions such as feature selection are automated, and the remaining manual steps hardly require domain-dependent knowledge. Furthermore, an extensive empirical evaluation demonstrates that the obtained objective functions yield more robustness. Learned objective functions enable fitting algorithms to determine the best model fit more accurately than with designed objective functions.