TY - GEN
T1 - Deep laplacian pyramid networks for fast and accurate super-resolution
AU - Lai, Wei Sheng
AU - Huang, Jia Bin
AU - Ahuja, Narendra
AU - Yang, Ming Hsuan
N1 - Funding Information:
This work is supported in part by the NSF CAREER Grant #1149783, gifts from Adobe and Nvidia. J.-B. Huang and N. Ahuja are supported in part by Office of Naval Research under Grant N00014-16-1-2314.
PY - 2017/11/6
Y1 - 2017/11/6
N2 - Convolutional neural networks have recently demonstrated high-quality reconstruction for single-image super-resolution. In this paper, we propose the Laplacian Pyramid Super-Resolution Network (LapSRN) to progressively reconstruct the sub-band residuals of high-resolution images. At each pyramid level, our model takes coarse-resolution feature maps as input, predicts the high-frequency residuals, and uses transposed convolutions for upsampling to the finer level. Our method does not require the bicubic interpolation as the pre-processing step and thus dramatically reduces the computational complexity. We train the proposed LapSRN with deep supervision using a robust Charbonnier loss function and achieve high-quality reconstruction. Furthermore, our network generates multi-scale predictions in one feed-forward pass through the progressive reconstruction, thereby facilitates resource-aware applications. Extensive quantitative and qualitative evaluations on benchmark datasets show that the proposed algorithm performs favorably against the state-of-the-art methods in terms of speed and accuracy.
AB - Convolutional neural networks have recently demonstrated high-quality reconstruction for single-image super-resolution. In this paper, we propose the Laplacian Pyramid Super-Resolution Network (LapSRN) to progressively reconstruct the sub-band residuals of high-resolution images. At each pyramid level, our model takes coarse-resolution feature maps as input, predicts the high-frequency residuals, and uses transposed convolutions for upsampling to the finer level. Our method does not require the bicubic interpolation as the pre-processing step and thus dramatically reduces the computational complexity. We train the proposed LapSRN with deep supervision using a robust Charbonnier loss function and achieve high-quality reconstruction. Furthermore, our network generates multi-scale predictions in one feed-forward pass through the progressive reconstruction, thereby facilitates resource-aware applications. Extensive quantitative and qualitative evaluations on benchmark datasets show that the proposed algorithm performs favorably against the state-of-the-art methods in terms of speed and accuracy.
UR - http://www.scopus.com/inward/record.url?scp=85041899955&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85041899955&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2017.618
DO - 10.1109/CVPR.2017.618
M3 - Conference contribution
AN - SCOPUS:85041899955
T3 - Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
SP - 5835
EP - 5843
BT - Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
Y2 - 21 July 2017 through 26 July 2017
ER -