About me

I currently work at Univrses as a Computer Vision and Deep Learning engineer. Mostly in and around SLAM and Visual Odometry problems.

I hold a double degree from KTH - Royal Institute of Technology in Stockholm and UPC - ETSETB TelecomBCN in Barcelona, with master's degrees in Electrical Eng. and in Telecommunications Eng. respectively.

I wrote my master's thesis at the National Institute of Informatics of Japan, in Tokyo. The goal of the thesis was to demonstrate the usefulness of deep multitask learning models for real-time deployment on embedded systems of solutions based in deep neural networks. The problem of building dynamic maps from UAV imagery via semantic segmentation and object detection was taken as an example. I was also involved in the creation of the Okutama-Action dataset for action detection from aerial view.

Previously, I spent a summer in Leuven as an intern at Intermodalics. I worked on an industrial mobile manipulation project which made extensive use of ROS and involved 3D vision, navigation and motion planning. During my first year of master's back home, I also had fun with the Big Data ecosystem as an intern at the Everis BigData Center of Excellence, analyzing tools, use cases, architectures and developing PoCs and demos.