TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Image Super-Resolution	BSD100 - 2x upscaling	SRFBN	PSNR	32.29	# 13
Image Super-Resolution	BSD100 - 3x upscaling	SRFBN	PSNR	29.24	# 10
Image Super-Resolution	BSD100 - 4x upscaling	SRFBN	PSNR	27.72	# 19
Image Super-Resolution	BSD100 - 4x upscaling	SRFBN	SSIM	0.7409	# 24
Image Super-Resolution	FFHQ 1024 x 1024 - 4x upscaling	SRFBN	FID	17.14	# 4
Image Super-Resolution	FFHQ 1024 x 1024 - 4x upscaling	SRFBN	MS-SSIM	0.931	# 6
Image Super-Resolution	FFHQ 1024 x 1024 - 4x upscaling	SRFBN	PSNR	27.90	# 5
Image Super-Resolution	FFHQ 1024 x 1024 - 4x upscaling	SRFBN	SSIM	0.822	# 5
Image Super-Resolution	FFHQ 256 x 256 - 4x upscaling	SRFBN	FID	132.59	# 5
Image Super-Resolution	FFHQ 256 x 256 - 4x upscaling	SRFBN	MS-SSIM	0.895	# 7
Image Super-Resolution	FFHQ 256 x 256 - 4x upscaling	SRFBN	PSNR	21.96	# 8
Image Super-Resolution	FFHQ 256 x 256 - 4x upscaling	SRFBN	SSIM	0.693	# 7
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	PSNR	29.577	# 4
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	SSIM	0.827	# 3
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	MS-SSIM	0.953	# 4
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	LLE	2.066	# 2
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	FED	0.0984	# 4
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	FID	20.032	# 6
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	LPIPS	0.2406	# 5
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	NIQE	13.901	# 7
Image Super-Resolution	Manga109 - 2x upscaling	SRFBN	PSNR	39.08	# 11
Image Super-Resolution	Manga109 - 3x upscaling	SRFBN	PSNR	34.18	# 8
Image Super-Resolution	Manga109 - 4x upscaling	SRFBN	PSNR	31.15	# 25
Image Super-Resolution	Manga109 - 4x upscaling	SRFBN	SSIM	0.9160	# 24
Image Super-Resolution	Set14 - 2x upscaling	SRFBN	PSNR	33.82	# 12
Image Super-Resolution	Set14 - 3x upscaling	SRFBN	PSNR	30.1	# 13
Image Super-Resolution	Set14 - 4x upscaling	SRFBN	PSNR	28.81	# 29
Image Super-Resolution	Set14 - 4x upscaling	SRFBN	SSIM	0.7868	# 32
Image Super-Resolution	Set5 - 2x upscaling	SRFBN	PSNR	38.11	# 14
Image Super-Resolution	Set5 - 3x upscaling	SRFBN	PSNR	34.70	# 11
Image Super-Resolution	Urban100 - 2x upscaling	SRFBN	PSNR	32.62	# 14
Image Super-Resolution	Urban100 - 3x upscaling	SRFBN	PSNR	28.73	# 12
Image Super-Resolution	Urban100 - 4x upscaling	SRFBN	PSNR	26.6	# 25
Image Super-Resolution	Urban100 - 4x upscaling	SRFBN	SSIM	0.8015	# 22

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-ffhq-1024-x-1024-4x)](https://paperswithcode.com/sota/image-super-resolution-on-ffhq-1024-x-1024-4x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-ffhq-512-x-512-4x)](https://paperswithcode.com/sota/image-super-resolution-on-ffhq-512-x-512-4x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-ffhq-256-x-256-4x)](https://paperswithcode.com/sota/image-super-resolution-on-ffhq-256-x-256-4x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-manga109-3x)](https://paperswithcode.com/sota/image-super-resolution-on-manga109-3x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-bsd100-3x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-bsd100-3x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-manga109-2x)](https://paperswithcode.com/sota/image-super-resolution-on-manga109-2x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-set5-3x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-set5-3x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-set14-2x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-set14-2x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-urban100-3x)](https://paperswithcode.com/sota/image-super-resolution-on-urban100-3x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-bsd100-2x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-bsd100-2x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-set14-3x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-set14-3x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-set5-2x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-set5-2x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-urban100-2x)](https://paperswithcode.com/sota/image-super-resolution-on-urban100-2x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-bsd100-4x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-bsd100-4x-upscaling?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-manga109-4x)](https://paperswithcode.com/sota/image-super-resolution-on-manga109-4x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-urban100-4x)](https://paperswithcode.com/sota/image-super-resolution-on-urban100-4x?p=feedback-network-for-image-super-resolution)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/feedback-network-for-image-super-resolution/image-super-resolution-on-set14-4x-upscaling)](https://paperswithcode.com/sota/image-super-resolution-on-set14-4x-upscaling?p=feedback-network-for-image-super-resolution)`

Feedback Network for Image Super-Resolution

CVPR 2019 · Zhen Li, Jinglei Yang, Zheng Liu, Xiaomin Yang, Gwanggil Jeon, Wei Wu ·

Recent advances in image super-resolution (SR) explored the power of deep learning to achieve a better reconstruction performance. However, the feedback mechanism, which commonly exists in human visual system, has not been fully exploited in existing deep learning based image SR methods. In this paper, we propose an image super-resolution feedback network (SRFBN) to refine low-level representations with high-level information. Specifically, we use hidden states in an RNN with constraints to achieve such feedback manner. A feedback block is designed to handle the feedback connections and to generate powerful high-level representations. The proposed SRFBN comes with a strong early reconstruction ability and can create the final high-resolution image step by step. In addition, we introduce a curriculum learning strategy to make the network well suitable for more complicated tasks, where the low-resolution images are corrupted by multiple types of degradation. Extensive experimental results demonstrate the superiority of the proposed SRFBN in comparison with the state-of-the-art methods. Code is avaliable at https://github.com/Paper99/SRFBN_CVPR19.

PDF Abstract CVPR 2019 PDF CVPR 2019 Abstract

Code

Add Remove Mark official

Paper99/SRFBN_CVPR19 official

550

turboLIU/SRFBN-tensorflow

tianbaochou/YOUKU-VSRE-2019-49th

JihyunLee9805/GMFN

Tasks

Add Remove

Image Super-Resolution

Super-Resolution

Datasets

FFHQ

BSD

DIV2K

Urban100

Set14

Set5

Manga109 BSD100

Results from the Paper

Edit

Ranked #4 on Image Super-Resolution on FFHQ 1024 x 1024 - 4x upscaling

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Image Super-Resolution	BSD100 - 2x upscaling	SRFBN	PSNR	32.29	# 13	Compare
Image Super-Resolution	BSD100 - 3x upscaling	SRFBN	PSNR	29.24	# 10	Compare
Image Super-Resolution	BSD100 - 4x upscaling	SRFBN	PSNR	27.72	# 19	Compare
Image Super-Resolution	BSD100 - 4x upscaling	SRFBN	SSIM	0.7409	# 24	Compare
Image Super-Resolution	FFHQ 1024 x 1024 - 4x upscaling	SRFBN	FID	17.14	# 4	Compare
			MS-SSIM	0.931	# 6	Compare
			PSNR	27.90	# 5	Compare
			SSIM	0.822	# 5	Compare
Image Super-Resolution	FFHQ 256 x 256 - 4x upscaling	SRFBN	FID	132.59	# 5	Compare
			MS-SSIM	0.895	# 7	Compare
			PSNR	21.96	# 8	Compare
			SSIM	0.693	# 7	Compare
Image Super-Resolution	FFHQ 512 x 512 - 4x upscaling	SRFBN	PSNR	29.577	# 4	Compare
			SSIM	0.827	# 3	Compare
			MS-SSIM	0.953	# 4	Compare
			LLE	2.066	# 2	Compare
			FED	0.0984	# 4	Compare
			FID	20.032	# 6	Compare
			LPIPS	0.2406	# 5	Compare
			NIQE	13.901	# 7	Compare
Image Super-Resolution	Manga109 - 2x upscaling	SRFBN	PSNR	39.08	# 11	Compare
Image Super-Resolution	Manga109 - 3x upscaling	SRFBN	PSNR	34.18	# 8	Compare
Image Super-Resolution	Manga109 - 4x upscaling	SRFBN	PSNR	31.15	# 25	Compare
Image Super-Resolution	Manga109 - 4x upscaling	SRFBN	SSIM	0.9160	# 24	Compare
Image Super-Resolution	Set14 - 2x upscaling	SRFBN	PSNR	33.82	# 12	Compare
Image Super-Resolution	Set14 - 3x upscaling	SRFBN	PSNR	30.1	# 13	Compare
Image Super-Resolution	Set14 - 4x upscaling	SRFBN	PSNR	28.81	# 29	Compare
Image Super-Resolution	Set14 - 4x upscaling	SRFBN	SSIM	0.7868	# 32	Compare
Image Super-Resolution	Set5 - 2x upscaling	SRFBN	PSNR	38.11	# 14	Compare
Image Super-Resolution	Set5 - 3x upscaling	SRFBN	PSNR	34.70	# 11	Compare
Image Super-Resolution	Urban100 - 2x upscaling	SRFBN	PSNR	32.62	# 14	Compare
Image Super-Resolution	Urban100 - 3x upscaling	SRFBN	PSNR	28.73	# 12	Compare
Image Super-Resolution	Urban100 - 4x upscaling	SRFBN	PSNR	26.6	# 25	Compare
Image Super-Resolution	Urban100 - 4x upscaling	SRFBN	SSIM	0.8015	# 22	Compare

Methods

Add Remove

No methods listed for this paper. Add relevant methods here

Edit Social Preview

Feedback Network for Image Super-Resolution

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove