A scalable method for surface reconstruction from noisy point clouds which allows the reconstruction of fine geometric details.
Sandro Lombardi, Martin R. Oswald, Marc Pollefeys
3DV 2020, Online
This is the long video for the paper:
Hi there! I'm Sandro Lombardi.
I’m a PhD student at the Computer Vision and Geometry Group at ETH Zurich, supervised by Prof. Marc Pollefeys.
Prior to starting my PhD, I’ve worked as a Research Assistant in collaboration with the ETH spin-off company Astrivis. In 2017, I obtained my MSc degree in computer science with the focus area of visual computing. I completed my BSc degree at ETH Zurich as well, after which I briefly worked as a software developer for the Balgrist University Hospital.
I love solving problems related to 3D reconstruction which are applicable to real world scenarios. My interests are diverse, be it reconstruction of dynamic, non-rigidly deforming scenes, reconstruction of body parts (e.g. faces or feet) or making reconstruction approaches scalable and applicable to large scenes.
Check out my projects or feel free to drop me a message!
Research in computer vision with focus on scalable meshing, human body representations and reconstruction of dynamic scenes.
Master's track in visual computing.
Major courses: Visual Computing, Information Security, Software Architecture and Engineering, Distributed Systems.
R&D in collaboration with Astrivis. Human foot reconstruction from mobile phones.
Assisting in development and improvement of a Computer Assisted Surgery Planning Application.
Development of Eclipse plugins and Eclipse-RCP-application for internal use.
This is a research project which I did in close collaboration with Astrivis. The goal was to fit a template 3D model of a human foot to a 3D scan, obtained with Astrivis’ 3D reconstruction pipeline on the smartphone.
This video shows the registration process. A template model is initialized close to the scanned foot. Afterwards the model is deformed non-rigidly and iteratively fitted to the points of the scan through some variant of the iterative closest point algorithm (ICP).