Intermediate Layer Optimization for Inverse Problems using Deep Generative Models

Joseph Dean, Giannis Daras, Alex Dimakis

Keywords: Inverse Problems, Inversion, Deep Generative Models

TL;DR: We propose Intermediate Layer Optimization, a novel optimization algorithm for solving inverse problems with deep generative models.

Abstract: We propose Intermediate Layer Optimization, a novel optimization algorithm for solving inverse problems with deep generative models. Instead of optimizing only over the initial latent code, we progressively change the input layer we optimize over, obtaining successively more expressive generators. We also experiment with different loss functions and utilize a perceptual loss combined with standard mean squared error. We empirically show that our approach outperforms the state-of-the-art inversion methods introduced in StyleGAN-2 and PULSE.

Community Implementations: [ 5 code implementations](https://www.catalyzex.com/paper/intermediate-layer-optimization-for-inverse/code)