Editing Out-of-domain GAN Inversion via Differential Activations

TL;DR

This paper introduces a novel GAN inversion framework that uses differential activation to improve out-of-domain image editing, effectively handling semantic changes and reducing ghosting effects for more faithful real-world image editing.

Contribution

It proposes a composition-decomposition paradigm with a differential activation module and a GAN prior based deghosting network to enhance out-of-domain GAN inversion and editing accuracy.

Findings

Outperforms state-of-the-art methods in qualitative evaluations

Achieves superior quantitative results in image editing tasks

Demonstrates robustness in both single and multi-attribute manipulations

Abstract

Despite the demonstrated editing capacity in the latent space of a pretrained GAN model, inverting real-world images is stuck in a dilemma that the reconstruction cannot be faithful to the original input. The main reason for this is that the distributions between training and real-world data are misaligned, and because of that, it is unstable of GAN inversion for real image editing. In this paper, we propose a novel GAN prior based editing framework to tackle the out-of-domain inversion problem with a composition-decomposition paradigm. In particular, during the phase of composition, we introduce a differential activation module for detecting semantic changes from a global perspective, \ie, the relative gap between the features of edited and unedited images. With the aid of the generated Diff-CAM mask, a coarse reconstruction can intuitively be composited by the paired original and edited images. In this way, the attribute-irrelevant regions can be survived in almost whole, while the quality of such an intermediate result is still limited by an unavoidable ghosting effect. Consequently, in the decomposition phase, we further present a GAN prior based deghosting network for separating the final fine edited image from the coarse reconstruction. Extensive experiments exhibit superiorities over the state-of-the-art methods, in terms of qualitative and quantitative evaluations. The robustness and flexibility of our method is also validated on both scenarios of single attribute and multi-attribute manipulations.