Marie-Paule Cani

Marie-Paule Cani is a Professor of Computer Science at Ecole Polytechnique (IP-Paris, France). Her research focuses on shape modelling and animation, with contributions to implicit surfaces, adaptive simulation methods and interactive natural scenes. She developed expressive ways to create them by combining sketching interfaces with models based on knowledge and learning. She received the Eurographics outstanding technical contributions award in 2011, was Siggraph Technical Papers Chair in 2017, joined the Siggraph Academy in 2019 and the French Academy of Sciences in 2020.

Creative AI: Turning 3D virtual worlds into experimental testbeds for scientists

Session Day and Time:
18 October 10:00am-11:00am

Visual representations are essential in science, as they facilitate intuition and ease communication. While the 2D sketches used for centuries recently evolved into 3D contents, the latter are generally not responsive: they do not allow scientists to interact with their models, change hypotheses on the fly and experiment their effects. In this talk, we explore the use of smart 3D environments as visual testbeds in science. We show that a combination of efficient procedural models expressing knowledge, expressive sculpting or sketching interfaces, and learning mechanisms, can help to reach this goal. We illustrate this through a variety of examples, from sculpting eroding mountains that express geological knowledge to helping prehistorians explore past ecosystems with flora and animated fauna.