Z
Zongze Wu
Researcher at Hebrew University of Jerusalem
Publications - 5
Citations - 402
Zongze Wu is an academic researcher from Hebrew University of Jerusalem. The author has contributed to research in topics: Real image & Computer science. The author has an hindex of 4, co-authored 5 publications receiving 75 citations.
Papers
More filters
Posted Content
StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generation
TL;DR: The latent style space of Style-GAN2, a state-of-the-art architecture for image generation, is explored and StyleSpace, the space of channel-wise style parameters, is shown to be significantly more disentangled than the other intermediate latent spaces explored by previous works.
Proceedings ArticleDOI
StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generation
TL;DR: In this paper, the authors explore and analyze the latent style space of Style-GAN2, a state-of-the-art architecture for image generation, using models pretrained on several different datasets.
Posted Content
StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery
TL;DR: In this article, a text-based interface for StyleGAN image manipulation is presented, which does not require human examination of the many degrees of freedom or annotated collection of images for each desired manipulation.
Proceedings Article
StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery
TL;DR: In this article, a text-based interface for StyleGAN image manipulation is presented, which does not require human examination of the many degrees of freedom or annotated collection of images for each desired manipulation.
Proceedings ArticleDOI
Fine-grained Foreground Retrieval via Teacher-Student Learning
TL;DR: In this paper, a self-supervised domain adaptation task is formulated for foreground image retrieval, where the source domain consists of foreground images and the target domain of background images, and foregrounds and background are effectively mapped into a common feature space, enabling retrieval of the foregrounds that are closest to the target background in that space.