Improved image inpainting network incorporating supervised attention module and cross-stage feature fusion

doi:10.11772/j.issn.1001-9081.2023020123

Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (2): 572-579.DOI: 10.11772/j.issn.1001-9081.2023020123

Special Issue: 多媒体计算与计算机仿真

• Multimedia computing and computer simulation • Previous Articles Next Articles

Improved image inpainting network incorporating supervised attention module and cross-stage feature fusion

Qiaoling HUANG¹, Bochuan ZHENG¹^,²(), Zicheng DING², Zedong WU²

^1.School of Mathematics & Information，China West Normal University，Nanchong Sichuan 637009，China
^2.School of Computer Science，China West Normal University，Nanchong Sichuan 637009，China

Received:2023-02-15 Revised:2023-05-08 Accepted:2023-05-11 Online:2023-08-14 Published:2024-02-10
Contact: Bochuan ZHENG
About author:HUANG Qiaoling， born in 1998， M. S. candidate. Her research interests include machine learning， deep learning， image inpainting.
DING Zicheng， born in 1999， M. S. candidate. His research interests include machine learning， deep learning， image inpainting.
WU Zedong， born in 2000， M. S. candidate. His research interests include machine learning， deep learning， object detection.
Supported by:
National Natural Science Foundation of China(62176217);Scientific Research Innovation Team Project of China West Normal University(KCXTD2022-3)

融合监督注意力模块和跨阶段特征融合的图像修复改进网络

黄巧玲¹, 郑伯川¹^,²(), 丁梓成², 吴泽东²

^1.西华师范大学数学与信息学院，四川南充 637009
^2.西华师范大学计算机学院，四川南充 637009

通讯作者: 郑伯川
作者简介:黄巧玲（1998—），女，四川绵阳人，硕士研究生，主要研究方向：机器学习、深度学习、图像修复
丁梓成（1999—），男，四川巴中人，硕士研究生，主要研究方向：机器学习、深度学习、图像修复
吴泽东（2000—），男，四川眉山人，硕士研究生，主要研究方向：机器学习、深度学习、目标检测。
基金资助:
国家自然科学基金资助项目(62176217);西华师范大学科研创新团队项目(KCXTD2022?3)

Abstract

Abstract:

Image inpainting techniques for non-regular missing regions are versatile but challenging. To address the problem that existing inpainting methods may produce artifacts， distorted structures， and blurred textures for high-resolution images， an improved image inpainting network， named Gconv_CS（Gated convolution based CSFF and SAM） incorporating Supervised Attention Module （SAM） and Cross-Stage Feature Fusion （CSFF） was proposed. In Gconv_CS， the SAM and CSFF were introduced to Cconv， a two-stage network model with gated convolution. SAM ensured the effectiveness of the incoming feature information to the next stage by providing a real image to supervise the output features of the previous stage. CSFF fused the features from the encoder-decoder of the previous stage and fed them to the encoder of the next stage to compensate for the loss of feature information in the previous stage. The experimental results show that， at a percentage of missing regions of 1% to 10%， compared with the baseline model Gconv， on CelebA-HQ dataset， Gconv_CS improved the Peak Signal-to-Noise Ratio （PSNR） and Structural SIMilarity index （SSIM） by 1.5% and 0.5% respectively， reduced Fréchet Inception Distance （FID） and L1 loss by 21.8% and 14.8% respectively； on Place2 dataset， the first two indicators increased by 26.7% and 0.8% respectively， and the latter two indicators decreased by 7.9% and 37.9% respectively. A good restoration effect was achieved when Gconv_CS was used to remove masks from a giant panda’s face.

Key words: image inpainting, two-stage network, Cross-Stage Feature Fusion (CSFF), Supervised Attention Module (SAM), gated convolution

摘要：

非规则缺失区域的图像修复技术用途广泛但具有挑战性。针对现有修复方法对高分辨率图像可能会产生伪影、扭曲结构和模糊纹理的问题，提出一种融合监督注意力模块（SAM）和跨阶段特征融合（CSFF）的图像修复改进网络（Gconv_CS）。在Gconv的两阶段网络模型上，引入了SAM与CSFF模块。SAM通过提供真实图像监督信号，监督上阶段输出特征，确保传入下阶段特征信息的有效性。CSFF将上阶段编码器-解码器的特征融合后送入下阶段的编码器，以弥补上阶段修复中特征信息的损失。实验结果表明，在缺失区域占比为1%~10%时，相较于基线模型Gconv，Gconv_CS在CelebA-HQ数据集上峰值信噪比（PSNR）和结构相似性指数（SSIM）分别提高了1.5%和0.5%，Fréchet起始距离（FID）和L1损失分别降低了21.8%、14.8%；在Place2数据集上，前2个指标分别提高了26.7%和0.8%，后2个指标分别降低了7.9%、37.9%。将Gconv_CS用于去除大熊猫面部遮挡物时，取得了较好的修复视觉效果。

关键词: 图像修复, 两阶段网络, 跨阶段特征融合, 监督注意力模块, 门控卷积

CLC Number:

TP391.41

Qiaoling HUANG, Bochuan ZHENG, Zicheng DING, Zedong WU. Improved image inpainting network incorporating supervised attention module and cross-stage feature fusion[J]. Journal of Computer Applications, 2024, 44(2): 572-579.

黄巧玲, 郑伯川, 丁梓成, 吴泽东. 融合监督注意力模块和跨阶段特征融合的图像修复改进网络[J]. 《计算机应用》唯一官方网站, 2024, 44(2): 572-579.

Figures/Tables 13

Fig. 1 Network structure of Gconv_CS

Tab. 1 Parameters of encoder-decoder in coarse network

层	参数	特征尺寸
Input	—	$512 × 512 × 3$
GConv	k=5； s=1； p=2	$512 × 512 × 48$
↓	k=3； s=2； p=1	$256 × 256 × 96$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
↓	k=3； s=2； p=1	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
D-GConv ×4	k=3； s=1； p=［2，4，8，16］；d=［2，4，8，16］	$128 × 128 × 192$
GConv	k=3； s=1； p=1	$128 × 128 × 192$
T-GConv	k=3； s=1； p=1	$256 × 256 × 96$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
T-GConv	k=3； s=1； p=1	$512 × 512 × 48$
GConv	k=3； s=1； p=1	$512 × 512 × 48$
SAM	—	$512 × 512 × 48$

Tab. 1 Parameters of encoder-decoder in coarse network

层	参数	特征尺寸
Input	—	$512 × 512 × 3$
GConv	k=5； s=1； p=2	$512 × 512 × 48$
↓	k=3； s=2； p=1	$256 × 256 × 96$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
↓	k=3； s=2； p=1	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
D-GConv ×4	k=3； s=1； p=［2，4，8，16］；d=［2，4，8，16］	$128 × 128 × 192$
GConv	k=3； s=1； p=1	$128 × 128 × 192$
T-GConv	k=3； s=1； p=1	$256 × 256 × 96$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
T-GConv	k=3； s=1； p=1	$512 × 512 × 48$
GConv	k=3； s=1； p=1	$512 × 512 × 48$
SAM	—	$512 × 512 × 48$

Tab. 2 Parameters of encoding branch structure in fine network

层	参数	特征尺寸
Input	—	$512 × 512 × 48$
↓	k=3； s=2； p=1	$256 × 256 × 48$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
↓	k=3； s=2； p=1	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
D-GConv ×4	k=3； s=1；p=［2，4，8，16］；d=［2，4，8，16］	$128 × 128 × 192$

Tab. 2 Parameters of encoding branch structure in fine network

层	参数	特征尺寸
Input	—	$512 × 512 × 48$
↓	k=3； s=2； p=1	$256 × 256 × 48$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
↓	k=3； s=2； p=1	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
D-GConv ×4	k=3； s=1；p=［2，4，8，16］；d=［2，4，8，16］	$128 × 128 × 192$

Tab. 3 Parameters of contextual attention encoding branch structure in fine network

层	参数	特征尺寸
Input	—	$512 × 512 × 48$
↓	k=3； s=2； p=1	$256 × 256 × 48$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
↓	k=3； s=2； p=1	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
Contextual Attention	—	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$

Tab. 3 Parameters of contextual attention encoding branch structure in fine network

层	参数	特征尺寸
Input	—	$512 × 512 × 48$
↓	k=3； s=2； p=1	$256 × 256 × 48$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
↓	k=3； s=2； p=1	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
Contextual Attention	—	$128 × 128 × 192$
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$

Tab. 4 Parameters of decoder in fine network

层	参数	特征尺寸
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
T-GConv	k=3； s=1； p=1	$256 × 256 × 96$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
T-GConv	k=3； s=1； p=1	$512 × 512 × 48$
GConv	k=3； s=1； p=1	$512 × 512 × 3$

Tab. 4 Parameters of decoder in fine network

层	参数	特征尺寸
GConv ×2	k=3； s=1； p=1	$128 × 128 × 192$
T-GConv	k=3； s=1； p=1	$256 × 256 × 96$
GConv	k=3； s=1； p=1	$256 × 256 × 96$
T-GConv	k=3； s=1； p=1	$512 × 512 × 48$
GConv	k=3； s=1； p=1	$512 × 512 × 3$

Tab. 5 Parameters of discriminator

层	参数	特征尺寸
Input	—	$512 × 512 × 4$
Conv	k=7； s=1； p=3	$512 × 512 × 64$
Conv	k=4； s=2； p=1	$256 × 256 × 128$
Conv	k=4； s=2； p=1	$128 × 128 × 256$
Conv	k=4； s=2； p=1	$64 × 64 × 256$
Conv	k=4； s=2； p=1	$32 × 32 × 256$
Conv	k=4； s=2； p=1	$16 × 16 × 1$

Tab. 5 Parameters of discriminator

层	参数	特征尺寸
Input	—	$512 × 512 × 4$
Conv	k=7； s=1； p=3	$512 × 512 × 64$
Conv	k=4； s=2； p=1	$256 × 256 × 128$
Conv	k=4； s=2； p=1	$128 × 128 × 256$
Conv	k=4； s=2； p=1	$64 × 64 × 256$
Conv	k=4； s=2； p=1	$32 × 32 × 256$
Conv	k=4； s=2； p=1	$16 × 16 × 1$

Fig. 2 Network structure of SAM

Fig. 3 Network structure of CSFF

Tab. 6 Ablation experiment result comparison

Gconv	CSFF	SAM	PSNR/dB	SSIM	FID	L1损失
√			37.801	0.987 3	0.876	0.002 74
√	√		38.004	0.987 7	0.782	0.002 60
√		√	38.021	0.987 9	0.775	0.002 59
√	√	√	38.364	0.992 4	0.684	0.002 32

Tab. 7 Quantitative experiment result comparison on CelebA-HQ dataset

Masks	模型	PSNR/dB	SSIM	FID	L1损失	Masks	模型	PSNR/dB	SSIM	FID	L1损失
［1%，10%）	CA	32.14	0.968	2.96	0.016 8	［30%，40%）	CA	21.58	0.843	18.62	0.045 6
	PEN	33.21	0.973	2.41	0.012 3		PEN	24.72	0.894	16.68	0.037 1
	PConv	35.65	0.986	1.58	0.008 6		PConv	24.62	0.905	12.69	0.032 5
	Gconv	37.80	0.987	0.87	0.002 7		Gconv	25.65	0.921	6.86	0.016 8
	Gconv_CS	38.36	0.992	0.68	0.002 3		Gconv_CS	26.97	0.925	6.14	0.016 2
［10%，20%）	CA	27.32	0.945	5.68	0.019 7	［40%，50%）	CA	19.98	0.796	23.45	0.067 5
	PEN	28.05	0.955	4.19	0.013 1		PEN	23.26	0.849	20.72	0.053 2
	PConv	30.01	0.962	3.68	0.011 2		PConv	23.96	0.886	18.56	0.043 4
	Gconv	31.22	0.968	2.05	0.005 9		Gconv	24.25	0.894	9.02	0.023 1
	Gconv_CS	32.68	0.970	1.92	0.005 8		Gconv_CS	25.04	0.897	8.55	0.021 3
［20%，30%）	CA	23.89	0.892	9.62	0.030 2	［50%，60%）	CA	17.85	0.711	58.35	0.086 6
	PEN	25.74	0.913	8.38	0.024 2		PEN	20.82	0.764	49.38	0.076 4
	PConv	27.22	0.926	6.28	0.016 8		PConv	20.96	0.808	38.25	0.065 2
	Gconv	27.58	0.945	3.93	0.007 1		Gconv	21.23	0.850	19.05	0.040 2
	Gconv_CS	29.41	0.947	3.73	0.006 8		Gconv_CS	21.71	0.854	17.32	0.039 3

Tab. 8 Quantitative experiment result comparison on Places2 dataset

Masks	模型	PSNR/dB	SSIM	FID	L1损失	Masks	模型	PSNR/dB	SSIM	FID	L1损失
［1%，10%）	CA	29.06	0.961	2.08	0.008 9	［30%，40%）	CA	18.90	0.783	18.46	0.051 9
	PEN	30.54	0.964	1.99	0.007 4		PEN	20.18	0.795	17.96	0.039 7
	PConv	31.04	0.971	1.86	0.006 8		PConv	20.44	0.820	16.33	0.034 3
	Gconv	32.25	0.970	1.51	0.006 6		Gconv	21.56	0.796	15.62	0.040 8
	Gconv_CS	33.11	0.978	1.39	0.004 1		Gconv_CS	22.36	0.861	34.26	0.031 4
［10%，20%）	CA	24.82	0.906	7.33	0.020 7	［40%，50%）	CA	16.84	0.721	56.89	0.070 7
	PEN	25.17	0.917	7.14	0.015 8		PEN	18.60	0.733	34.04	0.054 0
	PConv	26.75	0.928	6.85	0.012 8		PConv	18.86	0.762	28.63	0.046 2
	Gconv	27.05	0.921	5.85	0.015 5		Gconv	19.65	0.727	24.66	0.056 8
	Gconv_CS	27.96	0.944	4.12	0.010 2		Gconv_CS	20.94	0.819	22.35	0.044 6
［20%，30%）	CA	20.93	0.844	17.36	0.035 4	［50%，60%）	CA	15.11	0.668	82.67	0.101 1
	PEN	21.15	0.856	15.66	0.027 2		PEN	16.62	0.664	52.27	0.077 6
	PConv	22.59	0.875	13.83	0.024 2		PConv	16.85	0.667	47.09	0.077 4
	Gconv	23.84	0.860	12.40	0.027 3		Gconv	17.96	0.626	32.90	0.080 9
	Gconv_CS	24.62	0.903	34.26	0.022 3		Gconv_CS	18.98	0.766	31.35	0.076 2

Fig. 4 Visual comparison of five image inpainting models

Fig. 5 Display of giant panda face de-masking

References 33

1	BERTALMIO M， SAPIRO G， CASELLES V， et al. Image inpainting［C］// Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques. New York：ACM，2000： 417-424. 10.1145/344779.344972
2	BALLESTER C， BERTALMIO M， CASELLES V， et al. Filling-in by joint interpolation of vector fields and gray levels［J］. IEEE Transactions on Image Processing， 2001， 10（8）： 1200-1211. 10.1109/83.935036
3	TSCHUMPERLÉ D， DERICHE R. Vector-valued image regularization with PDEs： a common framework for different applications［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2005， 27（4）： 506-517. 10.1109/tpami.2005.87
4	BARNES C， SHECHTMAN E， FINKELSTEIN A， et al. PatchMatch： a randomized correspondence algorithm for structural image editing［J］. ACM Transactions on Graphics， 2009， 28（3）： Article No. 24. 10.1145/1531326.1531330
5	HUANG J-B， KANG S B， AHUJA N， et al. Image completion using planar structure guidance［J］. ACM Transactions on Graphics， 2014， 33（4）： No. 129. 10.1145/2601097.2601205
6	DING D， RAM S， RODRÍGUEZ J J. Image inpainting using nonlocal texture matching and nonlinear filtering［J］. IEEE Transactions on Image Processing， 2019， 28（4）： 1705-1719. 10.1109/tip.2018.2880681
7	PATHAK D， KRÄHENBÜHL P， DONAHUE J， et al. Context encoders： feature learning by inpainting［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE，2016： 2536-2544. 10.1109/cvpr.2016.278
8	GOODFELLOW I， POUGET-ABADIE J， MIRZA M， et al. Generative adversarial networks ［J］. Communications of the ACM， 2020， 63（11）： 139-144. 10.1145/3422622
9	LIZUKA S， SIMO-SERRA E， ISHIKAWA H. Globally and locally consistent image completion ［J］. ACM Transactions on Graphics， 2017， 36（4）： Article No. 107. 10.1145/3072959.3073659
10	LIU G， REDA F A， SHIH K J， et al. Image inpainting for irregular holes using partial convolutions［C］// Proceedings of the 15th European Conference on Computer Vision. Cham： Springer，2018： 89-105. 10.1007/978-3-030-01252-6_6
11	XIE C， LIU S， LI C， et al. Image inpainting with learnable bidirectional attention maps［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 8857-8866. 10.1109/iccv.2019.00895
12	ZENG Y， FU J， CHAO H， et al. Learning pyramid-context encoder network for high-quality image inpainting［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 1486-1494. 10.1109/cvpr.2019.00158
13	LIU H， JIANG B， SONG Y， et al. Rethinking image inpainting via a mutual encoder-decoder with feature equalizations［C］// Proceedings of the 16th European Conference on Computer Vision. Cham： Springer， 2020： 725-741. 10.1007/978-3-030-58536-5_43
14	LIAO L， XIAO J， WANG Z， et al. Guidance and evaluation： semantic-aware image inpainting for mixed scenes［C］// Proceedings of the 10th European Conference on Computer Vision. Cham： Springer， 2020： 683-700. 10.1007/978-3-030-58583-9_41
15	ZHANG R， QUAN W， WU B， et al. Pixel‐wise dense detector for image inpainting［J］. Computer Graphics Forum， 2020， 39（7）： 471-482. 10.1111/cgf.14160
16	WANG N， ZHANG Y， ZHANG L. Dynamic selection network for image inpainting［J］. IEEE Transactions on Image Processing， 2021， 30： 1784-1798. 10.1109/tip.2020.3048629
17	ZHU M， HE D， LI X， et al. Image inpainting by end-to-end cascaded refinement with mask awareness［J］. IEEE Transactions on Image Processing， 2021， 30： 4855-4866. 10.1109/tip.2021.3076310
18	YU J， LIN Z， YANG J， et al. Generative image inpainting with contextual attention［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE，2018： 5505-5514. 10.1109/cvpr.2018.00577
19	YU J， LIN Z， YANG J， et al. Free-form image inpainting with gated convolution［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE，2019： 4470-4479. 10.1109/iccv.2019.00457
20	ZHANG H， HU Z， LUO C， et al. Semantic image inpainting with progressive generative networks［C］// Proceedings of the 26th ACM International Conference on Multimedia. New York：ACM， 2018： 1939-1947. 10.1145/3240508.3240625
21	NAZERI K， NG E， JOSEPH T， et al. EdgeConnect： structure guided image inpainting using edge prediction［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision Workshops.Piscataway： IEEE， 2019： 3265-3274. 10.1109/iccvw.2019.00408
22	REN Y， YU X， ZHANG R， et al. StructureFlow： image inpainting via structure-aware appearance flow［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE，2019： 181-190. 10.1109/iccv.2019.00027
23	WU H， ZHOU J， LI Y. Deep generative model for image inpainting with local binary pattern learning and spatial attention［EB/OL］.［2023-02-01］. . 10.1109/tmm.2021.3111491
24	袁琳君，蒋旻，罗敦浪，等.基于生成对抗网络的人像修复［J］.计算机应用，2020，40（3）：842-846.
	YUAN L J， JIANG M， LUO D L， et al. Portrait inpainting based on generative adversarial networks ［J］. Journal of Computer Applications，2020，40（3）：842-846.
25	冯浪，张玲，张晓龙.基于扩张卷积的图像修复［J］.计算机应用，2020，40（3）：825-831. 10.11772/j.issn.1001-9081.2019081471
	FENG L， ZHANG L， ZHANG X L. Image inpainting based on dilated convolution［J］. Journal of Computer Applications，2020，40（3）：825-831. 10.11772/j.issn.1001-9081.2019081471
26	HOCHREITER S， SCHMIDHUBER J. Long short-term memory［J］. Neural Computation， 1997， 9（8）： 1735-1780. 10.1162/neco.1997.9.8.1735
27	GUO Z， CHEN Z， YU T， et al. Progressive image inpainting with full-resolution residual network［C］// Proceedings of the 27th ACM International Conference on Multimedia. New York：ACM， 2019： 2496-2504. 10.1145/3343031.3351022
28	LI J， HE F， ZHANG L， et al. Progressive reconstruction of visual structure for image inpainting［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 5961-5970. 10.1109/iccv.2019.00606
29	LI J， WANG N， ZHANG L， et al. Recurrent feature reasoning for image inpainting［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 7760-7768. 10.1109/cvpr42600.2020.00778
30	ZENG Y， LIN Z， YANG J， et al. High-resolution image inpainting with iterative confidence feedback and guided upsampling［C］// Proceedings of the 16th European Conference on Computer Vision. Cham： Springer， 2020： 1-17. 10.1007/978-3-030-58529-7_1
31	KARRAS T， AILA T， LAINE S， et al. Progressive growing of GANs for improved quality， stability， and variation［EB/OL］.［2023-02-01］. . 10.1109/cvpr42600.2020.00813
32	ZHOU B， LAPEDRIZA A， KHOSLA A， et al. Places： A 10 million image database for scene recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2018， 40（6）： 1452-1464. 10.1109/tpami.2017.2723009
33	ZAMIR S W， ARORA A， KHAN S， et al. Multi-stage progressive image restoration［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE，2021： 14816-14826. 10.1109/cvpr46437.2021.01458

[1]	Kaili DENG, Weibo WEI, Zhenkuan PAN. Industrial defect detection method with improved masked autoencoder [J]. Journal of Computer Applications, 2024, 44(8): 2595-2603.
[2]	Juxiang ZHOU, Jinsheng LIU, Jianhou GAN, Di WU, Zijie LI. Classroom speech emotion recognition method based on multi-scale temporal-aware network [J]. Journal of Computer Applications, 2024, 44(5): 1636-1643.
[3]	Gang CHEN, Yongwei LIAO, Zhenguo YANG, Wenying LIU. Image inpainting algorithm of multi-scale generative adversarial network based on multi-feature fusion [J]. Journal of Computer Applications, 2023, 43(2): 536-544.
[4]	Wanli SHEN, Yujin ZHANG, Wan HU. U-shaped feature pyramid network for image inpainting forensics [J]. Journal of Computer Applications, 2023, 43(2): 545-551.
[5]	WANG Xiao, WEI Jiawang, YUAN Yubo. Face image inpainting method based on circular fields of feature parts [J]. Journal of Computer Applications, 2020, 40(3): 847-853.
[6]	YUAN Linjun, JIANG Min, LUO Dunlang, JIANG Jiajun, GUO Jia. Portrait inpainting based on generative adversarial networks [J]. Journal of Computer Applications, 2020, 40(3): 842-846.
[7]	FENG Lang, ZHANG Ling, ZHANG Xiaolong. Image inpainting based on dilated convolution [J]. Journal of Computer Applications, 2020, 40(3): 825-831.
[8]	YANG Wenxia, WANG Meng, ZHANG Liang. Semantic face image inpainting based on U-Net with dense blocks [J]. Journal of Computer Applications, 2020, 40(12): 3651-3657.
[9]	YANG Wenxia, ZHANG Liang. Image inpainting model based on structure-texture decomposition and local total variation minimization [J]. Journal of Computer Applications, 2018, 38(8): 2386-2392.
[10]	YANG Wenxia, ZHANG Liang. Logarithmic function based non-local total variation image inpainting model [J]. Journal of Computer Applications, 2018, 38(6): 1784-1789.
[11]	MENG Hongyue, ZHAI Donghai, LI Mengxue, CAO Daming. Image inpainting algorithm based on pruning samples referring to four-neighborhood [J]. Journal of Computer Applications, 2018, 38(4): 1111-1116.
[12]	CAO Daming, ZHAI Donghai, MENG Hongyue, LI Mengxue, FENG Yan. Image inpainting algorithm based on priori constraints and statistics [J]. Journal of Computer Applications, 2018, 38(2): 533-538.
[13]	LIN Jinyong, DENG Dexiang, YAN Jia, LIN Xiaoying. Self-adaptive group based sparse representation for image inpainting [J]. Journal of Computer Applications, 2017, 37(4): 1169-1173.
[14]	ZOU Weigang, ZHOU Zhihui, WANG Yang. Image inpainting algorithm based on non-subsampled contourlet transform [J]. Journal of Computer Applications, 2017, 37(2): 553-558.
[15]	LI Mengxue, ZHAI Donghai, MENG Hongyue, CAO Daming. Image inpainting algorithm for partitioning feature subregions [J]. Journal of Computer Applications, 2017, 37(12): 3541-3546.

Improved image inpainting network incorporating supervised attention module and cross-stage feature fusion

融合监督注意力模块和跨阶段特征融合的图像修复改进网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 13

References 33

Related Articles 15

Recommended Articles

Metrics