Abstract
Most of learning-based super-resolution (SR) methods reconstructed high resolution (HR) image using a very deep convolutional neural network (CNN). CNN-based methods are used to process low resolution images, the predicted image is too smooth and some texture details are ignored. In order to improve the quality of the recovered high resolution image, wavelet transform is applied at the end of our model. Firstly, the prediction image is obtained through the CNN model; then, the predicted image and the ground truth image are processed into four frequency channels by wavelet transform; at last, four frequency channels and original image are considered when calculating loss. Compared with the original CNN model, the results of adding wavelet transform is obviously improved.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
