Mesh-Based Freeform Lens Optimization for Light Control

Abstract: 
Designing a freeform surface that reflects or refracts light to achieve a target distribution is a challenging inverse problem. In this talk, I will present two recent works on optimizing freeform lenses represented as triangle meshes to create desired caustic patterns. The first work introduces a differentiable rendering model that calculates caustics from a mesh surface without the need for sampling. This model enables direct optimization based on the difference between the target and resulting caustic images, leading to single-sided freeform lens designs with caustics closing matching the target. In the second work, we propose a technique that simultaneously optimizes both sides of a freeform lens, using a bijective correspondence between the two sides to control light refraction paths. Our preliminary results are promising, suggesting new possibilities for freeform optics design. 

Bio:
Bailin Deng is a Senior Lecturer in the School of Computer Science and Informatics at Cardiff University, where he leads the Geometric Computing and Graphics Research Group. He holds a BEng in Computer Software and an MSc in Computer Science from Tsinghua University, as well as a PhD in Technical Mathematics from the Vienna University of Technology. Before joining Cardiff University in 2017, he worked as a postdoctoral researcher at the Computer Graphics and Geometry Laboratory at EPFL, and as a Lecturer in the School of Computer Science at the University of Hull. His research focuses on 3D geometry processing and the development of numerical algorithms, with applications in computational design, geometric modeling, and computer animation.