Guilherme G. Schardong
Introduction
I'm currently an Assistant Professor at Universidade Federal de Santa Maria (UFSM), campus Cachoeira do Sul. Until May 2026, I was a Research Scientist at Institute of Systems and Robotics, University of Coimbra, where I worked on projects with the Visual Information Security Team (VISTEAM). Up to January 2022, I was a research fellow at GALGOS/PUC-Rio, with experience on oil&gas industry from projects with PETROBRAS on well-testing and reservoir model inference. From 2020 to 2022 I was also a part-time post-doc at Visgraf/IMPA, where I worked on computer graphics and machine learning, with emphasis on implicit function reconstruction, morphing, and rendering.
Background
I got my Bachelor's degree in Computer Science from UFSM in 2012 and a Master's degree in Informatics in the same university, in 2014, advised by prof. Cesar Pozzer. In 2018 I got my D.Sc. in Informatics at PUC-Rio, as a member of GALGOS. From 2014 to 2016 I worked as a developer under prof. Waldemar Celes on the Reservoir Visualization group on Tecgraf/PUC-Rio Institute. In 2016 I started working with prof. Hélio Lopes (being advised by him since 2015) and was co-advised by prof. Simone Barbosa during my doctorate. My latest position was as a Research Scientist at the Institute for Systems an Robotics, University of Coimbra at the VISTEAM group, led by prof Nuno Gonçalves. I also colaborate on a regular basis with prof. Luiz Velho and Tiago Novello from Visgraf/IMPA.
Research
My current research interests involve mainly AI for Graphics, Vision and Implicit Neural Representations.
Journal Papers
- Democratizing interactivity: An overview of interfaces for multimedia machine learning [Paper link] [Info]
- Geometric implicit neural representations for signed distance functions [Paper link] [Info]
- MR-Net: Multiresolution sinusoidal neural networks [Project page] [Paper link] [Info]
- Exploring Differential Geometry in Neural Implicits [Project page] [Code repository] [Paper link] [ArXiv link] [Info]
Conference Papers
- FLOWING: Implicit Neural Flows for Structure-Preserving Morphing [Project page] [Code repository] [ArXiv link] [Info]
- RiemStega: Covariance-Based Loss for Print-Proof Transmission of Data in Images [Paper link] [Info]
- INTERACT-NET: An Interactive Interface for Multimedia Machine Learning [Paper link] [Info]
- Neural Implicit Morphing of Face Images [Project page] [Code repository] [Paper link] [ArXiv link] [Info]
- How to train your (neural) dragon [Project page] [Code repository] [Paper link] [Info]
- High-Resolution Detection of Earth Structural Heterogeneities from Seismic Amplitudes using Convolutional Neural Networks with Attention layers [Paper link] [ArXiv link] [Info]
- Neural Implicit Surface Evolution [Project page] [Code repository] [Paper link] [ArXiv link] [Info]
- Multiresolution Neural Networks for Imaging [Project page] [Paper link] [ArXiv link] [Info]
Patents
- Method for detecting anomalies in the well and oil reservoir system using artificial intelligence [Patent link] [Info]