Instead of direct con- catenation, multiple adaptive spatial feature fusion (ASFF) blocks are adopted for combining the features from guid- ance and degraded ...

Overview of our ASFFNet. WLS is adopted to select the optimal guidance from multiple exemplars, MLS and AdaIN are leveraged for spatial alignment and ...

To address these issues, this paper suggests to enhance blind face restoration performance by utilizing multi-exemplar images and adaptive fusion of features ...

This paper suggests to enhance blind face restoration performance by utilizing multi-exemplar images and adaptive fusion of features from guidance and degraded ...

Instead of direct con- catenation, multiple adaptive spatial feature fusion (ASFF) blocks are adopted for combining the features from guid- ance and degraded ...

Jun 1, 2020 · To enhance blind face restoration performance by utilizing multi-exemplar images and adaptive fusion of features from guidance and degraded ...

Enhanced Blind Face Restoration with Multi-Exemplar Images and Adaptive Spatial Feature Fusion (CVPR 2020) This is a re-implemented version of our ASFFNet.

Blind face restoration aims at recovering high-quality faces from the low-quality counterparts suffering from unknown degradation.

Feb 4, 2022 · In this article, we propose a multi-prior collaboration network (MPCNet) to seamlessly integrate the advantages of generative priors and face-specific geometry ...

This work proposes a new method by first learning a GAN for high-quality face image generation and embedding it into a U-shaped DNN as a prior decoder.