Figure - uploaded by Marc Teva Law
Content may be subject to copyright.
Download
View publication
Figure A.2: Truncated boundary of positive semidefinite cone in S 2 plotted in R 3 as (a, b, c). The boundary of S 2 + is the set of parameters (a, b, c) that satisfy a ≥ 0, c ≥ 0, ac = b 2 , it represents the set of symmetric PSD matrices that are not full rank. The interior of the cone is the set of positive definites matrices (S 2 ++ ).

Figure A.2: Truncated boundary of positive semidefinite cone in S 2 plotted in R 3 as (a, b, c). The boundary of S 2 + is the set of parameters (a, b, c) that satisfy a ≥ 0, c ≥ 0, ac = b 2 , it represents the set of symmetric PSD matrices that are not full rank. The interior of the cone is the set of positive definites matrices (S 2 ++ ).

Source publication
Figure 1.1: An illustration of the visual Bag-of-Words (BoW)...
Figure 1.2: BoW pipeline for classification
Figure 1.3: State-of-the-art results since 2006 on the Caltech 101...
Figure 1.4: High-level attributes describe categories of objects (here...
+10 Figure 1.7: LMNN: schematic illustration of one input's neighborhood...
Distance Metric Learning for Image and Webpage Comparison
Article
Full-text available
  • Jan 2015
This thesis focuses on distance metric learning for image and webpage comparison. Distance metrics are used in many machine learning and computer vision contexts such as k-nearest neighbors classification, clustering, support vector machine, information/image retrieval, visualization etc. In this thesis, we focus on Mahalanobis-like distance metric...
Cite
Download full-text