Structure-Preserving Image Super-resolution (SPSR)

Paper

Yukai Shi, Keze Wang, Chongyu Chen, Li Xu and Liang Lin. Structure-Preserving Image Super-resolution via Contextualized Multi-task Learning. To appear in IEEE Transactions on Mulitmedia (TMM), 2017. Paper

Downloads

Super-resolved images from various datasets: Set 5 Set14 BSD200

Source Code: Download

Efficiency

The efficiency analysis for the scaling factor of 3 on the Set5 dataset. The evaluation platform is a high performance desktop (CPU 4.0GHz, 32GB, GTX 1080). Our proposed SPSR is written in TensorFlow and fully optimized by the Factorized CNN

Performance

Quantitative comparisons among different methods in terms of PSNR (dB), in which the underline indicates the second place and bold face represents the first place.

Quantitative comparisons among different methods in terms of SSIM, in which the underline indicates the second place and bold face represents the first place.

Visual comparison on the “Zebra” image from Set14 (factor 3), where the PSNR and SSIM are separated by “/”.

Visual comparisons on the “Butterfly” image from Set5 (factor 4), where the PSNR and SSIM are separated by “/”.

Reference

A+Radu Timofte, Vincent De Smet, and Luc Van Gool, “A+: Adjusted anchored neighborhood regression for fast super-resolution,” in Asian Conference on Computer Vision. Springer, 2014, pp. 111–126.

SRCNNChao Dong, Chen Change Loy, Kaiming He, and Xiaoou Tang, “Learning a deep convolutional network for image super-resolution,” in Computer Vision–ECCV 2014, pp. 184–199. Springer, 2014.

SRFSamuel Schulter, Christian Leistner, and Horst Bischof, “Fast and accurate image upscaling with super-resolution forests,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 3791–3799.

FSRCNNChao Dong, Chen Change Loy, and Xiaoou Tang, “Accelerating the super-resolution convolutional neural network,” in European Conference on Computer Vision. Springer, 2016, pp. 391–407.

SCNZhaowen Wang, Ding Liu, Jianchao Yang, Wei Han, and Thomas Huang, “Deep networks for image super-resolution with sparse prior,” arXiv preprint arXiv:1507.08905, 2015.

ShCNNJimmy SJ. Ren, Li Xu, Qiong Yan, and Wenxiu Sun, “Shepard convolutional neural networks,” in Advances in Neural Information Processing Systems, 2015.

Factorized CNNMin Wang, Baoyuan Liu and Hassan Foroosh, “Factorized Convolutional Neural Networks,” in ICML, 2016.