Fast, Accurate, and, Lightweight Super-Resolution with Cascading Residual Network

By Namhyuk Ahn, Byungkon Kang, Kyung-ah Sohn. 2018

1,424 citations recorded by the source corpus.

In recent years, deep learning methods have been successfully applied to single-image super-resolution tasks. Despite their great performances, deep learning methods cannot be easily applied to real-world applications due to the requirement of heavy computation. In this paper, we address this issue by proposing an accurate and lightweight deep network for image super-resolution. In detail, we design an architecture that implements a cascading mechanism upon a residual network. We also present variant models of the proposed cascading residual network to further improve efficiency. Our extensive experiments show that even with much fewer parameters and operations, our models achieve performance comparable to that of state-of-the-art methods.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Hierarchical Neural Architecture Search for Single Image Super-Resolution (2020, neutral): “898 CARN [33] 222.”
Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search (2019, methodology): “3: The model FALSR-A (the one with best visual effects) comparable to CARN.”
Pre-Trained Image Processing Transformer (2020, neutral): “[2] and Lim et al .”
Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search (2019, background): “Most of the hottest research focuses on non-linear mapping process (Dong et al., 2014; Lim et al., 2017; Ahn et al., 2018; Haris et al., 2018) where the final performance matters most.”
Efficient Transformer for Single Image Super-Resolution (2021, methodology): “1) Objective Evaluation: In TABLE I, we compare our ESRT with 13 state-of-the-art lightweight SR models, including SRCNN [14], FSRCNN [15], VDSR [16], DRCN [3], LapSRN [43], DRRN [2], MemNet [4], EDSR-baseline [17], SRMDNF [44], CARN [7], FALSR [5], IMDN [11], and LatticeNet [23].”
Deep Learning for Image Super-Resolution: A Survey (2019, methodology): “[77] feature similarity MS-SSIM [74] multi-scale SSIM IFC [75] information ﬁdelity criterion PSNR peak signal-to-noise ratio IQA image quality assessment SSIM [62] structural similarity MOS mean opinion score VIF [76] visual information ﬁdelity MSSIM [62] mean SSIM BN [140] batch normalization GAN [”
Efficient Residual Dense Block Search for Image Super-Resolution (2019, neutral): “8493 CARN-M (Ahn et al. 2018) 46.”
Lightweight Image Super-Resolution with Information Multi-distillation Network (2019, methodology): “Scale LapSRN [14] IDN [11] EDSR-b [18] CARN [2] IMDN ×2 112K 175K 341K 157K 173K ×3 76K 75K 172K 90K 78K ×4 76K 51K 122K 76K 45K”
Densely Residual Laplacian Super-Resolution (2019, methodology): “To optimize speed and the number of parameters CARN [8] employed group convolutions.”
NTIRE 2020 Challenge on Real-World Image Super-Resolution: Methods and Results (2020, background): “search topic for decades [33, 17, 56, 65, 63, 69, 70, 71, 61, 13, 28, 64, 14, 15, 38, 40, 43, 16, 2, 3, 26, 29, 21] due to its many applications.”
Deep Learning for Single Image Super-Resolution: A Brief Review (2018, methodology): “3) Number of composite multiply-accumulate operations for measuring computational efficiency (Mult&Adds): Since operations in NNs for SISR are mainly multiplications with additions, we use Mult&Adds in CARN [105] to measure computation, assuming that the desired SR is 720p.”
Lightweight Feature Fusion Network for Single Image Super-Resolution (2019, background): “CARN-M [8] adopt group convolution to attack a trade-off between computation and performance of the model.”
Meta-Transfer Learning for Zero-Shot Super-Resolution (2020, background): “Dataset Bicubic CARN [2] RCAN [45] ZSSR [34] MZSR (1) MZSR (10)”
Efficient Long-Range Attention Network for Image Super-resolution (2022, methodology): “We first compare our ELAN-light with state-of-the-art light-weight SR models, including CNN-based models CARN[2], IMDN [18], LAPAR-A[25], Lat-ticeNet[37], and transformer-based models ESRT [36] and SwinIR-light[28].”
LatticeNet: Towards Lightweight Image Super-Resolution with Lattice Block (2020, methodology): “The LR images are obtained in the same way as [2, 36].”
Learned Image Downscaling for Upscaling Using Content Adaptive Resampler (2019, methodology): “SR models in the ‘Bicubic’ group include the SRCNN [28], VDSR [29], DRRN [55], MemNet [56], DnCNN [57], LapSRN [38], ZSSR [58], CARN [59], SRRAM [60].”
MADNet: A Fast and Lightweight Network for Single-Image Super Resolution (2020, methodology): “Furthermore, this approach limits real-world applications [1].”