TY - GEN
T1 - InsetGAN for Full-Body Image Generation
AU - Fruhstuck, Anna
AU - Singh, Krishna Kumar
AU - Shechtman, Eli
AU - Mitra, Niloy J.
AU - Wonka, Peter
AU - Lu, Jingwan
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - While GANs can produce photo-realistic images in ideal conditions for certain domains, the generation of full-body human images remains difficult due to the diversity of identities, hairstyles, clothing, and the variance in pose. In-stead of modeling this complex domain with a single GAN, we propose a novel method to combine multiple pretrained GANs, where one GAN generates a global canvas (e.g., human body) and a set of specialized GANs, or insets, focus on different parts (e.g., faces, shoes) that can be seamlessly inserted onto the global canvas. We model the problem as jointly exploring the respective latent spaces such that the generated images can be combined, by inserting the parts from the specialized generators onto the global canvas, without introducing seams. We demonstrate the setup by combining a full body GAN with a dedicated high-quality face GAN to produce plausible-looking humans. We evalu-ate our results with quantitative metrics and user studies.
AB - While GANs can produce photo-realistic images in ideal conditions for certain domains, the generation of full-body human images remains difficult due to the diversity of identities, hairstyles, clothing, and the variance in pose. In-stead of modeling this complex domain with a single GAN, we propose a novel method to combine multiple pretrained GANs, where one GAN generates a global canvas (e.g., human body) and a set of specialized GANs, or insets, focus on different parts (e.g., faces, shoes) that can be seamlessly inserted onto the global canvas. We model the problem as jointly exploring the respective latent spaces such that the generated images can be combined, by inserting the parts from the specialized generators onto the global canvas, without introducing seams. We demonstrate the setup by combining a full body GAN with a dedicated high-quality face GAN to produce plausible-looking humans. We evalu-ate our results with quantitative metrics and user studies.
KW - Image and video synthesis and generation
KW - Machine learning
UR - http://www.scopus.com/inward/record.url?scp=85134985852&partnerID=8YFLogxK
U2 - 10.1109/CVPR52688.2022.00757
DO - 10.1109/CVPR52688.2022.00757
M3 - Conference contribution
AN - SCOPUS:85134985852
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 7713
EP - 7722
BT - Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
PB - IEEE Computer Society
T2 - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
Y2 - 19 June 2022 through 24 June 2022
ER -