Balanced Two-Stage Residual Networks for Image Super-Resolution

By Yuchen Fan, Humphrey Shi, Jiahui Yu, Ding Liu, Wei Han, Haichao Yu, Zhangyang Wang, Xinchao Wang, Thomas S. Huang. 2017

108 citations recorded by the source corpus.

In this paper, balanced two-stage residual networks (BTSRN) are proposed for single image super-resolution. The deep residual design with constrained depth achieves the optimal balance between the accuracy and the speed for super-resolving images. The experiments show that the balanced two-stage structure, together with our lightweight two-layer PConv residual block design, achieves very promising results when considering both accuracy and speed. We evaluated our models on the New Trends in Image Restoration and Enhancement workshop and challenge on image super-resolution (NTIRE SR 2017). Our final model with only 10 residual blocks ranked among the best ones in terms of not only accuracy (6th among 20 final teams) but also speed (2nd among top 6 teams in terms of accuracy). The source code both for training and evaluation is available in https://github.com/ychfan/sr_ntire2017.

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

Pre-Trained Image Processing Transformer (2020, background): “A various excellent works are also proposed for the other tasks, such as denoising [68, 32, 37, 45, 24], dehazing [6, 46, 85, 80], deraining [36, 78, 62, 29, 74, 47], and debluring [67, 53, 23, 10].”
MADNet: A Fast and Lightweight Network for Single-Image Super Resolution (2020, background): “CNN-based SISR models [4], [5], [7], [8], [18], [24], [27], [56], [57] have shown dramatic improvements in recent years given their powerful nonlinear representation ability.”
Wide Activation for Efficient and Accurate Image Super-Resolution (2018, methodology): “Thus, in recent image SR networks [7, 19, 42], batch normalization is abandoned.”
Image Super-Resolution via Dual-State Recurrent Networks (2018, methodology): “Specifically, convolutional neural networks (CNNs) and residual learning [16] have been widely applied in much recent SR work [10, 12, 20, 21, 22, 25, 35, 38].”
NTIRE 2018 Challenge on Single Image Super-Resolution: Methods and Results (2018, unknown): “Secondly, UIUCIFP follows recent works [8, 21, 31] which remove the BN layer in the residual blocks and adopts weight normalization in their WDSR approach, although the introducing of weight normalization in training SR networks may not help that much, it enables the authors to use higher learning r”
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.”
A Deep Journey into Super-resolution (2019, methodology): “• BTSRN stands for balanced two-stage residual networks [45] for image super-resolution.”
Unsupervised Learning for Real-World Super-Resolution (2019, background): “While traditionally addressed with classical techniques [19, 12, 29, 36, 15], current approaches [9, 10, 22, 23, 26, 11, 2, 3, 14, 16] employ deep learning methodologies to train a mapping from LR to HR.”
Fast, Accurate, and, Lightweight Super-Resolution with Cascading Residual Network (2018, methodology): “Because of the richness of this dataset, recent SR models [4, 8, 26, 31] use DIV2K as well.”
NTIRE 2017 Challenge on Single Image Super-Resolution: Methods and Results (2017, unknown): “UIUC-IFP team The proposed balanced two-stage residual networks (BTSRN) [11] contains LR and HR stages with 6 and 4 residual blocks [13], resp.”