Open AccessDynamic Scene Deblurring Based on Semantic Information Supplement
Author(s) -
Yiming Liu,
Junhui Li,
Wenzhuo Huang,
Kang Tang,
Dan Xu
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1518/1/012069
Subject(s) - deblurring , computer science , generator (circuit theory) , artificial intelligence , image (mathematics) , feature (linguistics) , semantics (computer science) , information retrieval , pattern recognition (psychology) , data mining , image restoration , image processing , power (physics) , linguistics , physics , philosophy , quantum mechanics , programming language
To solve the problem of semantic information dilution in network propagation, a semantic information supplement mechanism (SIS) is proposed to improve the performance of dynamic scene deblurring algorithm. Based on GANs structure, our generator is to recycle the semantic information and features spanning across multiple receptive scales to restore a sharp image, when a blur image is given. What’s more, in order to better integrate semantic information with the latent-feature and solve the problem of training difficulty in very-deep network, we put forward a long and short skip-connection method. Extensive experiments show that our Semantic Information Supplement network (SIS-net) achieves both qualitative and quantitative improvements against state-of-the-art methods.