Automatic lens design based on differentiable ray-tracing

Abstract

We propose a fully differentiable optical design method enabled by curriculum learning. Preliminary results show that our framework is suitable to solve highly non-convex problems like cellphone lens design.