DeepSEE: Deep Disentangled Semantic Explorative Extreme Super-Resolution
This addresses the need for diverse and controllable super-resolution outputs in facial imaging, though it is incremental as it builds on existing SR methods by adding semantic exploration.
The paper tackles the ill-posed problem of super-resolution by proposing DeepSEE, a framework for explorative facial super-resolution that leverages semantic maps to control regions and appearance, achieving up to 32x magnification and enabling image manipulations.
Super-resolution (SR) is by definition ill-posed. There are infinitely many plausible high-resolution variants for a given low-resolution natural image. Most of the current literature aims at a single deterministic solution of either high reconstruction fidelity or photo-realistic perceptual quality. In this work, we propose an explorative facial super-resolution framework, DeepSEE, for Deep disentangled Semantic Explorative Extreme super-resolution. To the best of our knowledge, DeepSEE is the first method to leverage semantic maps for explorative super-resolution. In particular, it provides control of the semantic regions, their disentangled appearance and it allows a broad range of image manipulations. We validate DeepSEE on faces, for up to 32x magnification and exploration of the space of super-resolution. Our code and models are available at: https://mcbuehler.github.io/DeepSEE/