A Unified Framework to Super-Resolve Face Images of Varied Low Resolutions
This addresses the inefficiency of training multiple models for different resolutions in FSR, offering a more practical solution for applications like surveillance or image enhancement, though it is incremental as it builds on existing FSR methods.
The authors tackled the problem of face image super-resolution (FSR) requiring separate models for each input resolution by proposing a unified framework that trains once and handles varied low resolutions, achieving robust and state-of-the-art performance across a wide range.
The existing face image super-resolution (FSR) algorithms usually train a specific model for a specific low input resolution for optimal results. By contrast, we explore in this work a unified framework that is trained once and then used to super-resolve input face images of varied low resolutions. For that purpose, we propose a novel neural network architecture that is composed of three anchor auto-encoders, one feature weight regressor and a final image decoder. The three anchor auto-encoders are meant for optimal FSR for three pre-defined low input resolutions, or named anchor resolutions, respectively. An input face image of an arbitrary low resolution is firstly up-scaled to the target resolution by bi-cubic interpolation and then fed to the three auto-encoders in parallel. The three encoded anchor features are then fused with weights determined by the feature weight regressor. At last, the fused feature is sent to the final image decoder to derive the super-resolution result. As shown by experiments, the proposed algorithm achieves robust and state-of-the-art performance over a wide range of low input resolutions by a single framework. Code and models will be made available after the publication of this work.