Patch-based Methods in Data-Driven Geometry Processing

Speaker:  Daniele Baieri

January 22^nd, 2026 | 9:00 am
DEIB, Conference Room "E. Gatti" (Bld. 20)

Contact: Prof. Luca Magri

Abstract

On January 22^nd, 2026, at 9:00 am the seminar on "Patch-based Methods in Data-Driven Geometry Processing" will take place at DEIB Conference Room "Emilio Gatti" (Building 20).

The recent advancements of deep learning led to powerful models for image and text processing. On the other hand, methods for 3D geometric data failed to keep track with their Euclidean counterparts. In this talk, we will discuss how patches, i.e. local regions of 3D surfaces, may be employed to develop general geometric neural models. To support our argument, we will show examples of recent algorithms whose core principles are based on local geometric information.

Short Bio

Daniele Baieri is a PostDoc researcher, currently in the 3DiG group led by Prof. Simone Melzi, at the University of Milano-Bicocca. He received his Ph.D. from Sapienza University of Rome in 2025, where he was tutored by Prof. Emanuele Rodolà. Daniele's research is oriented towards smart, efficient tools supporting artists in 3D graphics production pipelines. His recent work involved several topics including neural representations, Gaussian splatting, and geometric deep learning.