EnerViS (Energy Autonomous Low Power Vision Systems) is a joint project between Trento and Maryland (US) to study and develop of a novel vision system, capable of sensing and describing the visual world it observes under physical constraints that include ultra-low power consumption, low maintenance cost and a small unobtrusive form-factor.
You are here
MaDelENA (Developing and Studying novel intelligent nanoMaterials and Devices towards Adaptive Electronics and Neuroscience Applications), via a strongly innovative approach combining materials research, novel hardware design and computation, aims at developing neuro-bio-inspired electronic systems based on elements mimicking the function of components of the nervous system.
ALPS (Autonomous ultra Low-Power vision Sensor for surveillance and monitoring) will develop an ultra low-power vision sensor able to work with a minimal energy budget for wireless sensor networks in surveillance applications.
A paper on low power vision will be published in the Journal of Solid-State Circuits:
N. Cottini, M. Gottardi, N. Massari, R. Passerone, and Z. Smilansky, "A 33 W 6464 Pixel Vision Sensor Embedding Robust Dynamic Background Subtraction for Event Detection and Scene Interpretation," Journal of Solid-State Circuits, 2013.
Vision Sensors are special devices which not only capture the target image but also try to gather extra information directly at the chip level, making them an ideal component for an ultra-low power system. SOI research unit is investigating several vision sensors with advanced on-chip processing capabilities enabling beyond state-of-the-art low power performance.