|
|
Patricia Súarez, H. V., Dario Carpio & Angel Sappa. (2023). Corn Kernel Classification From Few Training Samples. In journal Artificial Intelligence in Agriculture, Vol. 9, pp. 89–99.
|
|
|
|
Cristhian A. Aguilera, Angel D. Sappa, & Ricardo Toledo. (2017). Cross-Spectral Local Descriptors via Quadruplet Network. In Sensors Journal, Vol. 17, pp. 873.
|
|
|
|
Xavier Soria, A. S., Patricio Humanante, Arash Akbarinia. (2023). Dense extreme inception network for edge detection. Pattern Recognition, Vol. 139.
|
|
|
|
Ortiz J., Londono J., Novillo F., Ampuno A., & Chávez M. (2015). Determinación de Invariantes en Grandes Centros de Datos basados en Topología Fat-Tree. Revista Politécnica, Vol. 35, pp. 91–96.
Abstract: Durante los últimos años ha existido un fuerte incremento en el acceso a internet, causando que los centros de datos ( DC) deban adaptar dinámicamente su infraestructura de red de cara a enfrentar posibles problemas de congestión, la cual no siempre se da de forma oportuna. Ante esto, nuevas topologías de red se han propuesto en los últimos años, como una forma de brindar mejores condiciones para el manejo de tráfico interno, sin embargo es común que para el estudio de estas mejoras, se necesite recrear el comportamiento de un verdadero DC en modelos de simulación/emulación. Por lo tanto se vuelve esencial validar dichos modelos, de cara a obtener resultados coherentes con la realidad. Esta validación es posible por medio de la identificación de ciertas propiedades que se deducen a partir de las variables y los parámetros que describen la red, y que se mantienen en las topologías de los DC para diversos escenarios y/o configuraciones. Estas propiedades, conocidas como invariantes, son una expresión del funcionamiento de la red en ambientes reales, como por ejemplo la ruta más larga entre dos nodos o el número de enlaces mínimo que deben fallar antes de una pérdida de conectividad en alguno de los nodos de la red. En el presente trabajo se realiza la identificación, formulación y comprobación de dos invariantes para la topología Fat-Tree, utilizando como software emulador a mininet. Las conclusiones muestran resultados concordantes entre lo analítico y lo práctico.
|
|
|
|
Abel Rubio, W. A., Leandro González & Jonathan Aviles-Cedeno. (2023). Distributed Intelligence in Autonomous PEM Fuel Cell Control. Energies 2023, Vol. 16(Issue 12).
|
|
|
|
Santos, V., Sappa, A.D., Oliveira, M. & de la Escalera, A. (2021). Editorial: Special Issue on Autonomous Driving and Driver Assistance Systems – Some Main Trends. In Journal: Robotics and Autonomous Systems. (Article number 103832), Vol. 144.
|
|
|
|
Marta Diaz, Dennys Paillacho, & Cecilio Angulo. (2015). Evaluating Group-Robot Interaction in Crowded Public Spaces: A Week-Long Exploratory Study in the Wild with a Humanoid Robot Guiding Visitors Through a Science Museum. International Journal of Humanoid Robotics, Vol. 12.
Abstract: This paper describes an exploratory study on group interaction with a robot-guide in an open large-scale busy environment. For an entire week a humanoid robot was deployed in the popular Cosmocaixa Science Museum in Barcelona and guided hundreds of people through the museum facilities. The main goal of this experience is to study in the wild the episodes of the robot guiding visitors to a requested destination focusing on the group behavior during displacement. The walking behavior follow-me and the face to face communication in a populated environment are analyzed in terms of guide- visitors interaction, grouping patterns and spatial formations. Results from observational data show that the space configurations spontaneously formed by the robot guide and visitors walking together did not always meet the robot communicative and navigational requirements for successful guidance. Therefore additional verbal and nonverbal prompts must be considered to regulate effectively the walking together and follow-me behaviors. Finally, we discuss lessons learned and recommendations for robot’s spatial behavior in dense crowded scenarios.
|
|
|
|
Rosero Vasquez Shendry. (2020). Facial recognition: traditional methods vs. methods based on deep learning. Advances in Intelligent Systems and Computing – Information Technology and Systems Proceedings of ICITS 2020.615–625.
|
|
|
|
Mónica Villavicencio, & Alain Abran. (2011). Facts and Perceptions Regarding Software Measurement in Education and in Practice: Preliminary Results. Journal of Software Engineering and Application, , pp. 227–234.
Abstract: How is software measurement addressed in undergraduate and graduate programs in universities? Do organizations consider that the graduating students they hire have an adequate knowledge of software measurement? To answer these and related questions, a survey was administered to participants who attended the IWSM-MENSURA 2010 conference in Stuttgart, Germany. Forty-seven of the 69 conference participants (including software development practitioners, software measurement consultants, university professors, and graduate students) took part in the survey. The results indicate that software measurement topics are: A) covered mostly at the graduate level and not at the undergraduate level, and B) not mandatory. Graduate students and professors consider that, of the measurement topics covered in university curricula, specific topics, such as measures for the requirements phase, and measurement techniques and tools, receive more attention in the academic context. A common observation of the practitioners who participated in the survey was that students hired as new employees bring limited software measurement-related knowledge to their organizations. Discussion of the findings and directions for future research are presented.
|
|
|
|
Cristhian A. Aguilera, C. A., Cristóbal A. Navarro, & Angel D. Sappa. (2020). Fast CNN Stereo Depth Estimation through Embedded GPU Devices. Sensors 2020, Vol. 2020-June(11), pp. 1–13.
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.
|
|
