One GPU, Learning In Less Than A Day! Ultra-Fast Learning GAN, "Lightweight GAN"
3 main points
✔️ Skip-Layer Excitation and self-supervised Discriminator proposed to reduce parameters significantly
✔️ It is possible to train even with small amounts of data
✔️ Even a 1024×1024 image can be trained on a single GPU in a few hours
Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
written by Anonymous
(Submitted on 29 Sep 2020)
Comments: Accepted at ICLR2021
Subjects: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Comm
overview
Until now, GANs, especially those like StyleGANs, have required a considerable amount of time for training. Our new lightweight GAN introduces two modules, Skip-Layer Excitation and self-supervised Discriminator, which significantly reduces the number of parameters. As a result, we are able to train at an astonishingly low cost: high quality images, small amounts of data, one GPU, and a few hours!
Here are the resulting images shown at the beginning. Both images are 1024×1024 in size and are trained with one GPU and 1000 images. The training time is 20 hours for the landscape image and 10 hours for the face image.
To read more,
Please register with AI-SCHOLAR.
ORCategories related to this article