← Prev: Innovation Week $10,000 Coding.. Next: Hacky Hour broaches the 3rd Di.. →


Egocentric Activity Detection at CVPR 2018

Egocentric Activity Detection at CVPR 2018

SIH data scientist and PhD student Rafael Pôssas presented at the 2018 Conference on Computer Vision and Pattern Recognition in Salt Lake City, Utah. His work is applying reinforcement learning, convolution neural networks and reinforcement neural networks to Egocentric Activity Recognition in wearables, balancing the tradeoff between predictive accuracy and low energy usage. The technology aims to help ageing people maitain their independence for longer by helping them to track their activity.

Also while at the conference, Rafael, Fabio and Sheila got a new GPU from NVIDIA!

Rafael, Fabio and Sheila showing off their new GPU.