文献类型：期刊文献

英文题名：Correction: AGG: attention-based gated convolutional GAN with prior guidance for image inpainting (Neural Computing and Applications, (2024), 36, 20, (12589-12604), 10.1007/s00521-024-09785-w)

作者：Yu, Xiankang[1]; Dai, Lei[1]; Chen, Zhihua[1]; Sheng, Bin[2]

机构：[1] Department of Computer Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China; [2] Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China

年份：2024

卷号：36

期号：20

起止页码：12605

外文期刊名：Neural Computing and Applications

收录：EI(收录号：20242316189435)

语种：英文

摘要：In this article the graphics relating to Figs.5, 6, 7 and 8 captions had been interchanged; the figure(s) should have appeared as shown below. (Figure presented.) (Figure presented.) (Figure presented.) (Figure presented.) Comparison of performance ranking among AGG and state-of-the-art methods in terms of FID, LPIPS, PSNR, and SSIM at different mask ratios on CelebA-HQ and Places2. RePaint [25] and MCG [26] do not have the results on Places2, so we can not compare them in the figures Qualitative comparison on Paris StreetView at three mask ratios, corresponding to 10–20%, 20–40%, and 40–60%. Our method can obtain more details without any artifacts Qualitative comparison on CelebA-HQ at 40–60% mask ratio.AGG obtains a more clear representation in the contour line on the face Qualitative comparison on Places2 at 20–40% mask ratio. The first row uses a manual mask to remove the ropes in the image, and our method has the best semantic results The original article has been corrected. ? The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.