Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution

TLDR

Convolutional neural networks have recently demonstrated high‑quality reconstruction for single‑image super‑resolution. The authors propose the Laplacian Pyramid Super‑Resolution Network (LapSRN) to progressively reconstruct sub‑band residuals of high‑resolution images. LapSRN progressively reconstructs high‑frequency residuals at each pyramid level using coarse feature maps, transposed convolutions for upsampling, deep supervision with a Charbonnier loss, and omits bicubic pre‑processing to reduce computational complexity. The network produces multi‑scale predictions in a single forward pass, enabling resource‑aware applications, and outperforms state‑of‑the‑art methods in both speed and accuracy on benchmark datasets.

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.