Real time motion estimation using a neural architecture implemented on GPUs

This work describes a neural network based architecture that represents and estimates object motion in videos. This architecture addresses multiple computer vision tasks such as image segmentation, object representation or characterization, motion analysis and tracking. The use of a neural network architecture allows for the simultaneous estimation of global and local motion and the representation of deformable objects. This architecture also avoids the problem of finding corresponding features while tracking moving objects. Due to the parallel nature of neural networks, the architecture has been implemented on GPUs that allows the system to meet a set of requirements such as: time constraints management, robustness, high processing speed and re-configurability. Experiments are presented that demonstrate the validity of our architecture to solve problems of mobile agents tracking and motion analysis.

Performance Factors in Women's Team Handball: Physical and Physiological Aspects—A Review

Team handball is an Olympic sport played professionally in many European countries. Nevertheless, a scientific knowledge regarding women's elite team handball demands is limited. Thus, the purpose of this article was to review a series of studies (n = 33) on physical characteristics, physiological attributes, physical attributes, throwing velocity, and on-court performances of women's team handball players. Such empirical and practical information is essential to design and implement successful short-term and long-term training programs for women's team handball players. Our review revealed that (a) players that have a higher skill level are taller and have a higher fat-free mass; (b) players who are more aerobically resistant are at an advantage in international level women team handball; (c) strength and power exercises should be emphasized in conditioning programs, because they are associated with both sprint performance and throwing velocity; (d) speed drills should also be implemented in conditioning programs but after a decrease in physical training volume; (e) a time-motion analysis is an effective method of quantifying the demands of team handball and provides a conceptual framework for the specific physical preparation of players. According to our results, there are only few studies on on-court performance and time-motion analysis for women's team handball players, especially concerning acceleration profiles. More studies are needed to examine the effectiveness of different training programs of women's team handball players' physiological and physical attributes.

Vertical crustal motion along the Mediterranean and Black Sea coast derived from ocean altimetry and tide gauge data

Tide gauge (TG) data along the northern Mediterranean and Black Sea coasts are compared to the sea-surface height (SSH) anomaly obtained from ocean altimetry (TOPEX/Poseidon and ERS-1/2) for a period of nine years (1993–2001). The TG measures the SSH relative to the ground whereas the altimetry does so with respect to the geocentric reference frame; therefore their difference would be in principle a vertical ground motion of the TG sites, though there are different error sources for this estimate as is discussed in the paper. In this study we estimate such vertical ground motion, for each TG site, from the slope of the SSH time series of the (non-seasonal) difference between the TG record and the altimetry measurement at a point closest to the TG. Where possible, these estimates are further compared with those derived from nearby continuous Global Positioning System (GPS) data series. These results on vertical ground motion along the Mediterranean and Black Sea coasts provide useful source data for studying, contrasting, and constraining tectonic models of the region. For example, in the eastern coast of the Adriatic Sea and in the western coast of Greece, a general subsidence is observed which may be related to the Adriatic lithosphere subducting beneath the Eurasian plate along the Dinarides fault.

A robust and fast method for 6DoF motion estimation from generalized 3D data

Nowadays, there is an increasing number of robotic applications that need to act in real three-dimensional (3D) scenarios. In this paper we present a new mobile robotics orientated 3D registration method that improves previous Iterative Closest Points based solutions both in speed and accuracy. As an initial step, we perform a low cost computational method to obtain descriptions for 3D scenes planar surfaces. Then, from these descriptions we apply a force system in order to compute accurately and efficiently a six degrees of freedom egomotion. We describe the basis of our approach and demonstrate its validity with several experiments using different kinds of 3D sensors and different 3D real environments.