Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution


(32x, 16x, 8x, 4x and 2x SR from a single SR model)

Abstract
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.

Paper

Citation

Wei-Sheng Lai, Jia-Bin Huang, Narendra Ahuja, and Ming-Hsuan Yang, "Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution", in IEEE Conference on Computer Vision and Pattern Recognition, 2017.


Bibtex
@inproceeding{LapSRN,
    author    = {Lai, Wei-Sheng and Huang, Jia-Bin and Ahuja, Narendra and Yang, Ming-Hsuan}, 
    title     = {Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution}, 
    booktitle = {IEEE Conferene on Computer Vision and Pattern Recognition},
    year      = {2017}
}
Download

Training Datasets (86.2 MB)
T91, BSDS200, General100

Testing Datasets (302 MB)
Set5, Set14, BSDS100, Urban100, Manga109, Historical

Network Architecture

Results
Benchmark

Performance on Set14





References