Search for: All records | NSF Public Access

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

Contrastive Learning for Diverse Disentangled Foreground Generation

https://doi.org/10.1007/978-3-031-19787-1_19

Li, Yuheng ; Li, Yijun ; Lu, Jingwan ; Shechtman, Eli ; Lee, Yong Jae ; Singh, Krishna Kumar ( January 2022 , European Conference on Computer Vision (ECCV))

Full Text Available
Collaging Class-specific GANs for Semantic Image Synthesis

https://doi.org/10.1109/iccv48922.2021.01415

Li, Yuheng ; Li, Yijun ; Lu, Jingwan ; Shechtman, Eli ; Lee, Yong Jae ; Singh, Krishna Kumar ( October 2021 , International Conference on Computer Vision (ICCV))

We propose a new approach for high resolution semantic image synthesis. It consists of one base image generator and multiple class-specific generators. The base generator generates high quality images based on a segmentation map. To further improve the quality of different objects, we create a bank of Generative Adversarial Networks (GANs) by separately training class-specific models. This has several benefits including – dedicated weights for each class; centrally aligned data for each model; additional training data from other sources, potential of higher resolution and quality; and easy manipulation of a specific object in the scene. Experiments show that our approach can generate high quality images in high resolution while having flexibility of object-level control by using class-specific generators. Project page: https://yuheng-li.github.io/CollageGAN/
more » « less
Full Text Available
Few-shot Image Generation via Cross-domain Correspondence

https://doi.org/10.1109/cvpr46437.2021.01060

Ojha, Utkarsh ; Li, Yijun ; Lu, Jingwan ; Efros, Alexei A. ; Lee, Yong Jae ; Shechtman, Eli ; Zhang, Richard ( June 2021 , Conference on Computer Vision and Pattern Recognition (CVPR))

Training generative models, such as GANs, on a target domain containing limited examples (e.g., 10) can easily result in overfitting. In this work, we seek to utilize a large source domain for pretraining and transfer the diversity information from source to target. We propose to preserve the relative similarities and differences between instances in the source via a novel cross-domain distance consistency loss. To further reduce overfitting, we present an anchor-based strategy to encourage different levels of realism over different regions in the latent space. With extensive results in both photorealistic and non-photorealistic domains, we demonstrate qualitatively and quantitatively that our few-shot model automatically discovers correspondences between source and target domains and generates more diverse and realistic images than previous methods.
more » « less
Full Text Available