RESUMEN

In real-life scenarios, Human Activity Recognition (HAR) from video data is prone to occlusion of one or more body parts of the human subjects involved. Although it is common sense that the recognition of the majority of activities strongly depends on the motion of some body parts, which when occluded compromise the performance of recognition approaches, this problem is often underestimated in contemporary research works. Currently, training and evaluation is based on datasets that have been shot under laboratory (ideal) conditions, i.e. without any kind of occlusion. In this work, we propose an approach for HAR in the presence of partial occlusion, in cases wherein up to two body parts are involved. We assume that human motion is modeled using a set of 3D skeletal joints and also that occluded body parts remain occluded during the whole duration of the activity. We solve this problem using regression, performed by a novel deep Convolutional Recurrent Neural Network (CRNN). Specifically, given a partially occluded skeleton, we attempt to reconstruct the missing information regarding the motion of its occluded part(s). We evaluate our approach using four publicly available human motion datasets. Our experimental results indicate a significant increase of performance, when compared to baseline approaches, wherein networks that have been trained using only nonoccluded or both occluded and nonoccluded samples are evaluated using occluded samples. To the best of our knowledge, this is the first research work that formulates and copes with the problem of HAR under occlusion as a regression task.

Asunto(s)

RESUMEN

Molybdenum oxide is used as a low-resistance anode interfacial layer in applications such as organic light emitting diodes and organic photovoltaics. However, little is known about the correlation between its stoichiometry and electronic properties, such as work function and occupied gap states. In addition, despite the fact that the knowledge of the exact oxide stoichiometry is of paramount importance, few studies have appeared in the literature discussing how this stoichiometry can be controlled to permit the desirable modification of the oxide's electronic structure. This work aims to investigate the beneficial role of hydrogenation (the incorporation of hydrogen within the oxide lattice) versus oxygen vacancy formation in tuning the electronic structure of molybdenum oxides while maintaining their high work function. A large improvement in the operational characteristics of both polymer light emitting devices and bulk heterojunction solar cells incorporating hydrogenated Mo oxides as hole injection/extraction layers was achieved as a result of favorable energy level alignment at the metal oxide/organic interface and enhanced charge transport through the formation of a large density of gap states near the Fermi level.