Abstract: IoT (Internet of Things) and AI (Artificial Intelligence) are becoming
support tools for several current technological solutions due to significant advancements of these areas. The development of the IoT in various technological fields has contributed to predicting the behavior of various systems such as mechanical, electronic, and control using sensor networks. On the other hand, deep learning architectures have achieved excellent results in complex tasks, where patterns have been extracted in time series. This study has reviewed the most efficient deep learning architectures for forecasting and obtaining trends over time, together with data produced by IoT sensors. In this way, it is proposed to contribute to applications in fields in which IoT is contributing a technological advance such as smart cities, industry 4.0, sustainable agriculture, or robotics. Among the architectures studied in this article related to the process of time series data we have: LSTM (Long Short-Term Memory) for its high precision in prediction and the ability to automatically process input sequences; CNN (Convolutional Neural Networks) mainly in human activity
recognition; hybrid architectures in which there is a convolutional layer for data pre-processing and RNN (Recurrent Neural Networks) for data fusion from different sensors and their subsequent classification; and stacked LSTM Autoencoders that extract the variables from time series in an unsupervised way without the need of manual data pre-processing.Finally, well-known technologies in natural language processing are also used in time series data prediction, such as the attention mechanism and embeddings obtaining promising results.

Abstract: Current CNN-based stereo depth estimation models can barely run under real-time
constraints on embedded graphic processing unit (GPU) devices. Moreover, state-of-the-art
evaluations usually do not consider model optimization techniques, being that it is unknown what is
the current potential on embedded GPU devices. In this work, we evaluate two state-of-the-art models
on three different embedded GPU devices, with and without optimization methods, presenting
performance results that illustrate the actual capabilities of embedded GPU devices for stereo depth
estimation. More importantly, based on our evaluation, we propose the use of a U-Net like architecture
for postprocessing the cost-volume, instead of a typical sequence of 3D convolutions, drastically
augmenting the runtime speed of current models. In our experiments, we achieve real-time inference
speed, in the range of 5–32 ms, for 1216  368 input stereo images on the Jetson TX2, Jetson Xavier,
and Jetson Nano embedded devices.

Abstract: Autonomous vehicles have a large number of on-board sensors, not only for providing coverage all around the vehicle, but also to ensure multi-modality in the observation of the scene. Because of this, it is not trivial to come up with a single, unique representation that feeds from the data given by all these sensors. We propose an algorithm which is capable of mapping texture collected from vision based sensors onto a geometric description of the scenario constructed from data provided by 3D sensors. The algorithm uses a constrained Delaunay triangulation to produce a mesh which is updated using a specially devised sequence of operations. These enforce a partial configuration of the mesh that avoids bad quality textures and ensures that there are no gaps in the texture. Results show that this algorithm is capable of producing fine quality textures.

Abstract: This paper presents a domain adaptation strategy to efficiently train network architectures for estimating the relative camera pose in multi-view scenarios. The network architectures are fed by a pair of simultaneously acquired
images, hence in order to improve the accuracy of the solutions, and due to the lack of large datasets with pairs of
overlapped images, a domain adaptation strategy is proposed. The domain adaptation strategy consists on transferring the knowledge learned from synthetic images to real-world scenarios. For this, the networks are firstly
trained using pairs of synthetic images, which are captured at the same time by a pair of cameras in a virtual environment; and then, the learned weights of the networks are transferred to the real-world case, where the networks are retrained with a few real images. Different virtual 3D scenarios are generated to evaluate the
relationship between the accuracy on the result and the similarity between virtual and real scenarios—similarity
on both geometry of the objects contained in the scene as well as relative pose between camera and objects in the
scene. Experimental results and comparisons are provided showing that the accuracy of all the evaluated networks for estimating the camera pose improves when the proposed domain adaptation strategy is used,
highlighting the importance on the similarity between virtual-real scenarios.

Abstract: El presente trabajo compara dos Implementaciones de Centrales Telefónicas VoIP de Código Abierto implementados sobre una Red Inalámbrica Amallada. El primero comprende la instalación de la PBX en un servidor tradicional y el segundo la instalación de una PBX en un enrutador inalámbrico. Nuestro objetivo es
determinar cuál de estos dos sistemas es superior en cuanto a calidad de servicio se refiere. Para determinar la mejor solución, realizamos un estudio técnico de los paquetes capturados durante diferentes pruebas, considerando parámetros como el ancho de banda, retardo y jitter. Nuestros métodos de análisis pueden ser utilizados para futuros trabajos con una mayor complejidad y número de enrutadores inalámbrico, así como establecer el grado de afectación y el comportamiento de las dos PBX cuando haya congestión en la red.