Abstract:Face hallucination is a domain-specific super-resolution problem that aims to generate a high-resolution (HR) face image from a low-resolution~(LR) input. In contrast to the existing patch-wise super-resolution models that divide a face image into regular patches and independently apply LR to HR mapping to each patch, we implement deep reinforcement learning and develop a novel attention-aware face hallucination (Attention-FH) framework, which recurrently learns to attend a sequence of patches and performs facial part enhancement by fully exploiting the global interdependency of the image. Specifically, our proposed framework incorporates two components: a recurrent policy network for dynamically specifying a new attended region at each time step based on the status of the super-resolved image and the past attended region sequence, and a local enhancement network for selected patch hallucination and global state updating. The Attention-FH model jointly learns the recurrent policy network and local enhancement network through maximizing a long-term reward that reflects the hallucination result with respect to the whole HR image. Extensive experiments demonstrate that our Attention-FH significantly outperforms the state-of-the-art methods on in-the-wild face images with large pose and illumination variations.

Position-based face hallucination method

Hallucinating Face by Position-Patch

Hallucinating Faces: Global Linear Modal Based Super-Resolution and Position Based Residue Compensation

Hallucinating Faces: LPH Super-Resolution and Neighbor Reconstruction for Residue Compensation.

An Example-Based Two-Step Face Hallucination Method through Coefficient Learning

A Two-Step Approach to Hallucinating Faces: Global Parametric Model and Local Nonparametric Model

MAP-Based Two-Step Approach to Hallucinating Faces

An Improved Two-Step Approach to Hallucinating Faces

A synthesis method for personalized 3D face reconstruction

Face Hallucination VIA Sparse Coding.

Face hallucination via position-based dictionaries coding in kernel feature space

Learning Adaptive Local Distance Metric For Face Hallucination

Face Hallucination by Learning Local Distance Metric.

Contextual-Field Supported Iterative Representation for Face Hallucination.

Locality-Constrained Double Low-Rank Representation for Effective Face Hallucination

High-fidelity 3D Face Reconstruction with Multi-Scale Details

Steerable pyramid-based face hallucination

Hallucinating facial images and features

C-Face: Using Compare Face on Face Hallucination for Low-Resolution Face Recognition

Identity-Preserving Face Hallucination via Deep Reinforcement Learning

Face Hallucination by Attentive Sequence Optimization with Reinforcement Learning