Light field image super-resolution network based on angular difference enhancement

Tianqi Lv; Yingchun Wu; Xianling Zhao

doi:10.12086/oee.2023.220185

Opto-Electronic Engineering, Volume. 50, Issue 2, 220185(2023)

Light field image super-resolution network based on angular difference enhancement

Tianqi Lv... Yingchun Wu* and Xianling Zhao |Show fewer author(s)

School of Electronic and Information Engineering, Taiyuan University of Science and Technology, Taiyuan, Shanxi 030024, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(17)

Fig. 1. 4D light field acquisition and rearrangement. (a) Biplanar representation model of light field; (b) Subaperture image array; (c) Macropixel array

Download full size

View in Article

Fig. 2. Overall network structure diagram

Download full size

View in Article

Fig. 3. Multi-branch residual block

Download full size

View in Article

Fig. 4. ADA module details. (a) ADA module data processing process; (b) Feature collection; (c) Offset acquisition in feature collection; (d) Update the central view; (e) Feature distribution; (f) Offset acquisition in feature distribution

Download full size

View in Article

Fig. 5. EADA module details. (a) EADA feature collection details; (b) EADA feature distribution details

Download full size

View in Article

Fig. 6. RFD module simplification. (a) RFD module details; (b) SRFD module details

Download full size

View in Article

Fig. 7. Visual contrast of the "Origami" scene with 2^× SR

Download full size

View in Article

Fig. 8. Visual contrast of the "Herbs" scene with 2^× SR

Download full size

View in Article

Fig. 9. Visual contrast of the "Bee" scene with 4^× SR

Download full size

View in Article

Fig. 10. Visual contrast of the "Lego Knights" scene with 4^× SR

Download full size

View in Article

Table 1. Five public light field datasets used in our experiment
View table
View in Article
Table 1. Five public light field datasets used in our experiment
数据集 EPFL^[13] HCInew^[14] HCIold^[15] INRIA^[16] STFgantry^[17] 总共
#训练 70 20 10 35 9 144
#测试 10 4 2 5 2 23

Table 2. PSNR/SSIM values achieved by different shallow feature extraction modules for 4×SR
View table
View in Article
Table 2. PSNR/SSIM values achieved by different shallow feature extraction modules for 4×SR
模型 EPFL^[13] HCInew^[14] HCIold^[15] INRIA^[16] STFgantry^[17]
无残差 25.26/0.8324 27.71/0.8517 32.58/0.9344 26.95/0.8867 26.09/0.8452
MBR 28.81/0.9190 31.30/0.9206 37.39/0.9725 30.81/0.9513 31.29/0.9511

Table 3. PSNR/SSIM values achieved by different deep feature extraction modules for 4^× SR

View table

View in Article

Table 3. PSNR/SSIM values achieved by different deep feature extraction modules for 4^× SR

模型	EPFL^[13]	HCInew^[14]	HCIold^[15]	INRIA^[16]	STFgantry^[17]
模型	EPFL^[13]	HCInew^[14]	HCIold^[15]	INRIA^[16]	STFgantry^[17]	ADA模块	28.77/0.9172	31.26/0.9198	37.41/0.9723	30.80/0.9507	31.17/0.9497
单支路EADA模块	28.78/0.9166	31.21/0.9186	37.31/0.9717	30.85/0.9502	31.03/0.9474
EADA模块	28.81/0.9190	31.30/0.9206	37.39/0.9725	30.81/0.9513	31.29/0.9511

Table 4. PSNR/SSIM values achieved by different feature fusion modules for 4^×SR
View table
View in Article
Table 4. PSNR/SSIM values achieved by different feature fusion modules for 4^×SR
模型 EPFL^[13] HCInew^[14] HCIold^[15] INRIA^[16] STFgantry^[17]
RFD模块 29.01/0.9183 31.32/0.9198 37.39/0.9718 31.08/0.9509 31.14/0.9499
SRFD模块 28.81/0.9190 31.30/0.9206 37.39/0.9725 30.81/0.9513 31.29/0.9511

Table 5. PSNR/SSIM values achieved by different methods for 2^× SR

View table

View in Article

Table 5. PSNR/SSIM values achieved by different methods for 2^× SR

超分方法	EPFL^[13]	HCInew^[14]	HCIold^[15]	INRIA^[16]	STFgantry^[17]	Average
超分方法	EPFL^[13]	HCInew^[14]	HCIold^[15]	INRIA^[16]	STFgantry^[17]	Average	EDSR^[20]	33.09/0.9631	34.83/0.9594	41.01/0.9875	34.97/0.9765	36.29/0.9819	36.04/0.9728
RCAN^[21]	33.16/0.9635	34.98/0.9602	41.05/0.9875	35.01/0.9769	36.33/0.9825	36.11/0.9741
ResLF^[8]	32.75/0.9672	36.07/0.9715	42.61/0.9922	34.57/0.9784	36.89/0.9873	36.58/0.9793
LFSSR^[7]	33.69/0.9748	36.86/0.9753	43.75/0.9939	35.27/0.9834	38.07/0.9902	37.53/0.9835
LF-InterNet^[4]	34.14/0.9761	37.28/0.9769	44.45/0.9945	35.80/0.9846	38.72/0.9916	38.08/0.9847
LF-DFnet^[5]	34.44/0.9766	37.44/0.9786	44.23/0.9943	36.36/0.9841	39.61/0.9935	38.42/0.9854
本文方法	34.58/0.9772	37.92/0.9796	44.84/0.9948	36.59/0.9854	40.11/0.9939	38.81/0.9862

Table 6. PSNR/SSIM values achieved by different methods for 4^× SR

View table

View in Article

Table 6. PSNR/SSIM values achieved by different methods for 4^× SR

超分方法	EPFL^[13]	HCInew^[14]	HCIold^[15]	INRIA^[16]	STFgantry^[17]	Average
超分方法	EPFL^[13]	HCInew^[14]	HCIold^[15]	INRIA^[16]	STFgantry^[17]	Average	EDSR^[20]	27.84/0.8858	29.60/0.8874	35.18/0.9538	29.66/0.9259	28.70/0.9075	30.20/0.9121
RCAN^[21]	27.88/0.8863	29.63/0.8880	35.20/0.9540	29.76/0.9273	28.90/0.9110	30.27/0.9133
ResLF^[8]	27.46/0.8899	29.92/0.9011	36.12/0.9651	29.64/0.9339	28.99/0.9214	30.43/0.9223
LFSSR^[7]	28.27/0.9080	30.72/0.9124	36.70/0.9690	30.31/0.9446	30.15/0.9385	31.23/0.9345
LF-InterNet^[4]	28.67/0.9143	30.98/0.9165	37.11/0.9715	30.64/0.9486	30.53/0.9426	31.59/0.9387
LF-DFnet^[5]	28.77/0.9165	31.23/0.9196	37.32/0.9718	30.83/0.9503	31.15/0.9494	31.86/0.9415
本文方法	28.81/0.9190	31.30/0.9206	37.39/0.9725	30.81/0.9513	31.29/0.9511	31.92/0.9429

Table 7. Comparisons of the number of parameters and FLOPs by different methods for 2^× SR and 4^× SR

View table

View in Article

Table 7. Comparisons of the number of parameters and FLOPs by different methods for 2^× SR and 4^× SR

Method	Parameters/M	FLOPs/G
Method	Parameters/M	FLOPs/G	EDSR^[20]	38.62/38.89	39.56×25/40.66×25
RCAN^[21]	15.31/15.36	15.59×25/15.65×25
ResLF^[8]	6.35/6.79	37.06/39.70
LFSSR^[7]	0.81/1.61	25.70/128.44
LF-InterNet^[4]	4.80/5.23	47.46/50.10
LF-DFnet^[5]	3.94/3.99	57.22/57.31
本文方法	12.74/12.80	238.92/240.51

Tools

Get Citation

Copy Citation Text

Tianqi Lv, Yingchun Wu, Xianling Zhao. Light field image super-resolution network based on angular difference enhancement[J]. Opto-Electronic Engineering, 2023, 50(2): 220185

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Article

Received: Jul. 28, 2022

Accepted: Oct. 21, 2022

Published Online: Apr. 13, 2023

The Author Email: Wu Yingchun (yingchunwu3030@foxmail.com)

DOI:10.12086/oee.2023.220185

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Five public light field datasets used in our experiment

Table 1. Five public light field datasets used in our experiment

Table 2. PSNR/SSIM values achieved by different shallow feature extraction modules for 4×SR

Table 2. PSNR/SSIM values achieved by different shallow feature extraction modules for 4×SR

Table 3. PSNR/SSIM values achieved by different deep feature extraction modules for 4× SR

Table 3. PSNR/SSIM values achieved by different deep feature extraction modules for 4× SR

Table 4. PSNR/SSIM values achieved by different feature fusion modules for 4×SR

Table 4. PSNR/SSIM values achieved by different feature fusion modules for 4×SR

Table 5. PSNR/SSIM values achieved by different methods for 2× SR

Table 5. PSNR/SSIM values achieved by different methods for 2× SR

Table 6. PSNR/SSIM values achieved by different methods for 4× SR

Table 6. PSNR/SSIM values achieved by different methods for 4× SR

Table 7. Comparisons of the number of parameters and FLOPs by different methods for 2× SR and 4× SR

Table 7. Comparisons of the number of parameters and FLOPs by different methods for 2× SR and 4× SR

微信扫一扫：分享

Table 3. PSNR/SSIM values achieved by different deep feature extraction modules for 4^× SR

Table 3. PSNR/SSIM values achieved by different deep feature extraction modules for 4^× SR

Table 4. PSNR/SSIM values achieved by different feature fusion modules for 4^×SR

Table 4. PSNR/SSIM values achieved by different feature fusion modules for 4^×SR

Table 5. PSNR/SSIM values achieved by different methods for 2^× SR

Table 5. PSNR/SSIM values achieved by different methods for 2^× SR

Table 6. PSNR/SSIM values achieved by different methods for 4^× SR

Table 6. PSNR/SSIM values achieved by different methods for 4^× SR

Table 7. Comparisons of the number of parameters and FLOPs by different methods for 2^× SR and 4^× SR

Table 7. Comparisons of the number of parameters and FLOPs by different methods for 2^× SR and 4^× SR