Face Hallucination by Attentive Sequence Optimization with Reinforcement Learning
Autor: | Liang Lin, Keze Wang, Guanbin Li, Qingxing Cao, Yukai Shi |
---|---|
Rok vydání: | 2019 |
Předmět: |
FOS: Computer and information sciences
Face hallucination Computer science Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION 02 engineering and technology Iterative reconstruction Facial recognition system Machine Learning Artificial Intelligence Image Processing Computer-Assisted 0202 electrical engineering electronic engineering information engineering Humans Reinforcement learning Image resolution Image restoration business.industry Applied Mathematics Pattern recognition Visualization Recurrent neural network Computational Theory and Mathematics Face Face (geometry) 020201 artificial intelligence & image processing Neural Networks Computer Computer Vision and Pattern Recognition Artificial intelligence business Algorithms Software |
DOI: | 10.48550/arxiv.1905.01509 |
Popis: | 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. Comment: To be published in TPAMI |
Databáze: | OpenAIRE |
Externí odkaz: |