In this paper, we propose an improved method for single image super-resolution on the basis of residual learning and convolutional sparse coding (CSC). The key idea of it is to first perform a CSC-based decomposition on the input so that it can be split into two predefined parts: the smooth and residual components. Then, the extracted components are individually mapped according to their own characteristics, rather than directly perform a mapping from the original input. Specifically, we place more emphases on the residual one as it is much important to our task, while the smooth one is just propagated to the final output for providing a quick reference. Accordingly, the final architecture of our method conceptually integrates all the above steps into a completely end-to-end trainable deep network. Extensive experimental results indicate that our proposed method outperforms many state-of-the-art methods in terms of both visual fidelity and objective evaluation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.