AI Paper Atlas

Research map / Image generation / GANs

Dual Contrastive Loss and Attention for GANs

Ning Yu, Guilin Liu, Aysegul Dundar, Andrew Tao, Bryan Catanzaro, Larry S. Davis, Mario Fritz

ICCV 2021 · Image generation › GANs · approach: Large-scale GAN latent analysis · 58 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace non-saturating loss with dual contrastive loss (ablation)StyleGAN2 · FFHQ, Bedroom, Church, Horse, CLEVRFID: FFHQ 4.86 -> 3.98; Bedroom 4.01 -> 3.86; Church 4.54 -> 3.73; Horse 3.91 -> 3.70; CLEVR 9.62 -> 6.06Improved
add dual contrastive loss to other backbones (ablation)SNGAN · CLEVRFID: 29.19 -> 18.23Improved
add dual contrastive loss to other backbones (ablation)StyleGAN · CLEVRFID: 12.43 -> 8.96Improved
replace non-saturating loss with dual contrastive loss: discriminator feature distinguishability (ablation)StyleGAN2 · FFHQFDDF: 245. -> 377. (FDDF)Improved
add SAN self-attention in the generator (30k subsets) (ablation)StyleGAN2 · BedroomFID: 19.33 -> 16.36Improved
add DFN attention in the generator (ablation)StyleGAN2 · BedroomFID: 19.33 -> 26.86Hurt
add self-attention in the generator (large-scale)StyleGAN2 · FFHQ vs BedroomFID: FFHQ 4.86 -> 5.13; Bedroom 4.01 -> 3.48Mixed
add self-attention in the discriminator (ablation)StyleGAN2 · CelebA, Animal FaceFID: CelebA 9.84 -> 10.49; Animal Face 36.55 -> 42.41Hurt
add reference-attention in the discriminator (30k subsets) (ablation)StyleGAN2 · BedroomFID: 19.33 -> 8.32Improved
add reference-attention in the discriminator at larger data size (100K) (ablation)StyleGAN2 · CelebA 128x128, 100K imagesFID: 5.61 -> 6.86Hurt
add reference-attention in the discriminator at 1K images (ablation)StyleGAN2 · Bedroom 128x128, 1K imagesFID: 230.40 -> 79.81Improved
combine reference-attention discriminator with dual contrastive loss (ablation)StyleGAN2 w/ self-attn-G ref-attn-D · Bedroom (30k)FID: 8.32 -> 12.84Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy