相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Proceedings Paper
Computer Science, Artificial Intelligence
Hila Chefer et al.
Summary: The article distinguishes between style transfer and essence transfer, proposing a novel method that combines a Sty1eGAN generator and a semantic encoder to achieve both identity preservation and high-level feature transfer. Through extensive experiments, the superiority of the proposed method for essence transfer over existing methods is demonstrated.
COMPUTER VISION, ECCV 2022, PT XIII
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Jiayu Xiao et al.
Summary: This paper proposes a relaxed spatial structural alignment (RSSA) method to train a generative adversarial network (GAN) with limited data. By designing a cross-domain spatial structural consistency loss, it aligns the spatial structural information between synthesis image pairs of the source and target domains. Experimental results demonstrate that the proposed method outperforms existing approaches in few shot setting.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Shuai Yang et al.
Summary: This paper introduces a novel DualStyleGAN for high-resolution portrait style transfer. Compared to traditional methods, DualStyleGAN allows for more flexible control of styles in the original face domain and the artistic portrait domain, enabling high-quality portrait style transfer.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Article
Computer Science, Software Engineering
Omer Tov et al.
Summary: This paper studies the latent space of the StyleGAN unconditional generator and identifies tradeoffs between distortion-editability and distortion-perception within the latent space. Two principles for designing encoders to control the proximity of inversions to the regions originally trained on StyleGAN are suggested. By qualitatively and quantitatively evaluating on challenging domains like cars and horses, the inversion method presented here, along with common editing techniques, achieves superior real-image editing quality with only a minor drop in reconstruction accuracy.
ACM TRANSACTIONS ON GRAPHICS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Songhua Liu et al.
Summary: The study introduces a novel attention and normalization module, named Adaptive Attention Normalization (AdaAttN), which adaptively performs attentive normalization on a per-point basis.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Gihyun Kwon et al.
Summary: The research introduces a novel hierarchical adaptive Diagonal spatial ATtention (DAT) layers for disentangling spatial contents and styles in image generative models. This allows for more flexible control of spatial features in the generated images.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Tao Yang et al.
Summary: By embedding a prior GAN into a U-shaped DNN, the GPEN method can generate high-quality face images and achieve superior results in restoring severely degraded face images in the wild through fine-tuning with synthesized low-quality images. This approach can control the face structure, details, and background, producing visually realistic restoration results.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Mahmoud Afifi et al.
Summary: This paper introduces HistoGAN, a color histogram-based method for controlling the colors of GAN-generated images. By modifying the StyleGAN architecture effectively, the authors propose a way to control the colors of GAN-generated images based on target color histogram features. In addition, they demonstrate how HistoGAN can be expanded for recoloring real images with an unsupervised approach called ReHistoGAN.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Hyunsu Kim et al.
Summary: StyleMapGAN introduces a new method for editing images, enhancing the accuracy of encoder embedding through spatially variant modulation, and outperforming state-of-the-art models in various image manipulation tasks.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Rameen Abdal et al.
Summary: This research introduces an unsupervised segmentation framework for StyleGAN generated objects, splitting the generator into foreground and background networks to generate soft segmentation masks for foreground objects. Comparable results were achieved against state-of-the-art supervised segmentation networks on multiple object classes, with a clear improvement shown in qualitative and quantitative metrics compared to the best unsupervised segmentation approach.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Article
Chemistry, Multidisciplinary
Tin Kramberger et al.
APPLIED SCIENCES-BASEL
(2020)
Article
Computer Science, Artificial Intelligence
Xiaogang Wang et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2009)
