Intermediate Layer Optimization for Inverse Problems using Deep Generative Models
This addresses image reconstruction challenges in computer vision, offering a novel optimization approach for inverse problems.
The paper tackles inverse problems like inpainting and super-resolution by proposing Intermediate Layer Optimization (ILO), which progressively optimizes latent codes in deep generative models, improving error bounds and outperforming state-of-the-art methods such as StyleGAN-2 and PULSE.
We propose Intermediate Layer Optimization (ILO), 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 obtaining successively more expressive generators. To explore the higher dimensional spaces, our method searches for latent codes that lie within a small $l_1$ ball around the manifold induced by the previous layer. Our theoretical analysis shows that by keeping the radius of the ball relatively small, we can improve the established error bound for compressed sensing with deep generative models. We empirically show that our approach outperforms state-of-the-art methods introduced in StyleGAN-2 and PULSE for a wide range of inverse problems including inpainting, denoising, super-resolution and compressed sensing.