Eruptive history of the Krakatau volcanic complex, Indonesia

[EN] The 1883 eruption of Krakatau is one of the best known volcanic events in the world, although it was not the largest, nor the deadliest of known eruptions. However, the eruption happened in a critical moment (just after the first global telegraph network was established) and in a strategic place (the Sunda Straits were a naval traffic hot spot at that time). The lecture will explore these events in some detail before presenting an outline on ongoing multidisciplinary efforts to unravel the past and present day plumbing systems of the 1883 eruption and that of the active Anak Krakatau cone. A mid- and a lower-crustal magma storage level exist beneath the volcano, placing significant emphasis on magma-crust interaction in the uppermost, sediment-rich crust. This final aspect shares similarities with the 2011/2012 El Hierro eruption, highlighting the relevance of the interaction between ascending magmas and marine deposits that oceanic magmas have to pass. At Krakatau, shallow-level crustal contamination offers a possible explanation for the explosive nature of the 1883 eruption and also for those of the presently active Anak Krakatau edifice and helps constrain location, style and processes of subvolcanic magma storage.

Evaluation of LBP and HOG descriptors for clothing attribute description

[EN]In this work an experimental study about the capability of the LBP, HOG descriptors and color for clothing attribute classification is presented. Two different variants of the LBP descriptor are considered, the original LBP and the uniform LBP. Two classifiers, Linear SVM and Random Forest, have been included in the comparison because they have been frequently used in clothing attributes classification. The experiments are carried out with a public available dataset, the clothing attribute dataset, that has 26 attributes in total. The obtained accuracies are over 75% in most cases, reaching 80% for the necktie or sleeve length attributes.

An analysis of automatic gender classification

[EN]Different researches suggest that inner facial features are not the only discriminative features for tasks such as person identification or gender classification. Indeed, they have shown an influence of features which are part of the local face context, such as hair, on these tasks. However, object-centered approaches which ignore local context dominate the research in computational vision based facial analysis. In this paper, we performed an analysis to study which areas and which resolutions are diagnostic for the gender classification problem. We first demonstrate the importance of contextual features in human observers for gender classification using a psychophysical ”bubbles” technique.