T
Terrance E. Boult
Researcher at University of Colorado Colorado Springs
Publications - 285
Citations - 13190
Terrance E. Boult is an academic researcher from University of Colorado Colorado Springs. The author has contributed to research in topics: Facial recognition system & Biometrics. The author has an hindex of 51, co-authored 274 publications receiving 10853 citations. Previous affiliations of Terrance E. Boult include Columbia University & King Abdulaziz University.
Papers
More filters
Proceedings ArticleDOI
Towards Open Set Deep Networks
TL;DR: The proposed OpenMax model significantly outperforms open set recognition accuracy of basic deep networks as well as deep networks with thresholding of SoftMax probabilities, and it is proved that the OpenMax concept provides bounded open space risk, thereby formally providing anopen set recognition solution.
Journal ArticleDOI
Toward Open Set Recognition
TL;DR: This paper explores the nature of open set recognition and formalizes its definition as a constrained minimization problem, and introduces a novel “1-vs-set machine,” which sculpts a decision space from the marginal distances of a 1-class or binary SVM with a linear kernel.
Journal ArticleDOI
Probability Models for Open Set Recognition
TL;DR: The general idea of open space risk limiting classification is extended to accommodate non-linear classifiers in a multiclass setting and a new open set recognition model called compact abating probability (CAP), where the probability of class membership decreases in value as points move from known data toward open space.
Journal ArticleDOI
Constraining object features using a polarization reflectance model
TL;DR: The authors present a polarization reflectance model that uses the Fresnel reflection coefficients, which accurately predicts the magnitudes of polarization components of reflected light, and all the polarization-based methods presented follow from this model.
Journal ArticleDOI
Separation of Reflection Components Using Color and Polarization
TL;DR: A technique is developed for separating the specular and diffuse components of reflection from images that can handle highlights on surfaces with substantial texture, smoothly varying diffuse reflectance, and varying material properties.