Research map / Image generation / GANs
DA-GAN: Instance-Level Image Translation by Deep Attention Generative Adversarial Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use DA-GAN rather than UNIT | SVHN-to-MNIST domain adaptation · SVHN to MNIST | accuracy: 90.53 % -> 93.60 % | Helped |
| Use DA-GAN rather than StackGAN | CUB-200-2011 text-to-image · CUB-200-2011 | Inception score: 3.7 ± 0.4 -> 5.6± 0.4 | Helped |
| Use DA-GAN rather than StackGAN | CUB-200-2011 text-to-image · CUB-200-2011 | ♯miss: 36.0 -> 19.0 | Helped |
| Use DA-GAN rather than CycleGAN | Object configuration · Object configuration | Top-1 acc: 62.1 % -> 88.9 % | Helped |
| Add 10K DA-GAN pose-morphed training samples | Bird fine-grained classification · Bird fine-grained classification | accuracy: 79.0 % -> 81.6 % | Helped |
| Add the deep attention encoder (DAE) (ablation) | DA-GAN · MNIST translated samples | accuracy: 90.2 % -> 94.3 % | Helped |
| Add the consistency loss (ablation) | DA-GAN · MNIST translated samples | accuracy: 79.8% -> 94.3 % | Helped |
| Add the symmetry loss (ablation) | DA-GAN · MNIST translated samples | accuracy: 90.6% -> 94.3 % | Helped |
| Add discriminator D2 (ablation) | DA-GAN · MNIST translated samples | accuracy: 88.2% -> 94.3 % | Helped |
Closest papers by meaning
- InstaGAN: Instance-aware Image-to-Image Translation (ICLR 2019 · similarity 0.91)
- Unsupervised Image-to-Image Translation With Generative Prior (CVPR 2022 · similarity 0.90)
- Self-Supervised Dense Consistency Regularization for Image-to-Image Translation (CVPR 2022 · similarity 0.90)
- TraVeLGAN: Image-To-Image Translation by Transformation Vector Learning (CVPR 2019 · similarity 0.90)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.89)
- Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned Manifold (AAAI 2024 · similarity 0.88)
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.