Xiaojie Li1, Yongpeng Ren1, Hongping Ren1, Canghong Shi2, Xian Zhang1, Lutao Wang1, Imran Mumtaz3, Xi Wu1
CMC-Computers, Materials & Continua, Vol.71, No.3, pp. 5021-5037, 2022, DOI:10.32604/cmc.2022.023071
- 14 January 2022
Abstract Recently, deep learning-based image outpainting has made greatly notable improvements in computer vision field. However, due to the lack of fully extracting image information, the existing methods often generate unnatural and blurry outpainting results in most cases. To solve this issue, we propose a perceptual image outpainting method, which effectively takes the advantage of low-level feature fusion and multi-patch discriminator. Specifically, we first fuse the texture information in the low-level feature map of encoder, and simultaneously incorporate these aggregated features reusability with semantic (or structural) information of deep feature map such that we could utilize More >