Type
Conference PaperAuthors
Alharbi, Yazeed
Wonka, Peter

KAUST Department
Computer Science ProgramComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Visual Computing Center (VCC)
Date
2020-08-05Preprint Posting Date
2020-04-26Online Publication Date
2020-08-05Print Publication Date
2020-06Permanent link to this record
http://hdl.handle.net/10754/662872
Metadata
Show full item recordAbstract
We explore different design choices for injecting noise into generative adversarial networks (GANs) with the goal of disentangling the latent space. Instead of traditional approaches, we propose feeding multiple noise codes through separate fully-connected layers respectively. The aim is restricting the influence of each noise code to specific parts of the generated image. We show that disentanglement in the first layer of the generator network leads to disentanglement in the generated image. Through a grid-based structure, we achieve several aspects of disentanglement without complicating the network architecture and without requiring labels. We achieve spatial disentanglement, scale-space disentanglement, and disentanglement of the foreground object from the background style allowing fine-grained control over the generated images. Examples include changing facial expressions in face images, changing beak length in bird images, and changing car dimensions in car images. This empirically leads to better disentanglement scores than state-of-the-art methods on the FFHQ dataset.Citation
Alharbi, Y., & Wonka, P. (2020). Disentangled Image Generation Through Structured Noise Injection. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). doi:10.1109/cvpr42600.2020.00518Conference/Event name
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)ISBN
978-1-7281-7169-2arXiv
2004.12411Additional Links
https://ieeexplore.ieee.org/document/9157760/https://ieeexplore.ieee.org/document/9157760/
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9157760
ae974a485f413a2113503eed53cd6c53
10.1109/CVPR42600.2020.00518