In this paper we provide an approach on sports analysis using Deep learning techniques. As part of a current project, the volleyball's basic reception technique has been divided into temporal phases. We performed an evaluation over our own labelled dataset consisting in 14814 frames from 69 videos depicting the desired reception technique. A model based on the YOLO algorithm was trained to locate the player region and trim the frames. Two time fusion methods over the frames wereproposed and evaluated with CNN models which were created based on the ResNet models and a transfer learning approach was used to train them. The results show that these models were able of classifying the frames with their corresponding phase with an accuracy of 92.21% in our best model. Also it can be seen that the RGB merging method shown in this paper helps to slightly improve the performance of the models. Fu...

This paper presents the development and validation of a tridimensional 7-DOF human body model for the representation and study of closed kinetic chain exercises (CKCE) performed with the feet fixed in space, i.e. low posture exercises. The biomechanical model, a link-segment model, is based on an Euler-Lagrange formulation and employs a generalized joint coordinate system. A top-down mechanical analysis provides an estimation of the internal joint moments, along with the vertical ground reaction forces, using kinematical data collected by inertial sensors. The model is validated by correlating estimated ground reaction forces to kinetic experimental data from force plates. Pearson correlation coefficients were calculated for four CKCE types (150 trials in total). In all cases, a median correlation r > 0.90 was found, hence proving that the proposed model is quite satisfactory for CKCE mo...