Led by lab members Christian Shewmake and Sophia Sanborn, the Geometric Intelligence Lab (PI: Nina Miolane) receives a 1.2M$ NSF grant to build deep vision models capable of discovering and modeling the hierarchical symmetries of the visual world, in collaboration with UC Berkeley (PI: Bruno Olshausen) and the University of Michigan (PI: Stella Yu).
With this NSF grant, titled Lie Group Representation Learning for Vision, the team is set to provide a new framework that incorporates the notions of hierarchical transformations and symmetry discovery into the geometric deep learning paradigm. The team will continue its efforts building and animating the NeurReps community.