Research map / Image generation / GANs
Image Generation from Sketch Constraint Using Contextual GAN
Yongyi Lu, Shangzhe Wu, Yu-Wing Tai, Chi-Keung Tang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use contextual GAN instead of pix2pix on good sketches | sketch-to-face generation · CelebA test (1000 images) | SSIM: 0.9012 -> 0.8856 | Hurt |
| use contextual GAN instead of pix2pix | sketch-to-face generation · CelebA test (1000 images) | face verification accuracy: 99.69 -> 99.80 | Improved |
| use contextual GAN instead of Context Encoder | sketch-to-face generation · CelebA test (1000 images) | SSIM: 0.5477 -> 0.8856 | Improved |
| use contextual GAN instead of Context Encoder | sketch-to-face generation · CelebA test (1000 images) | face verification accuracy: 97.19 -> 99.80 | Improved |
| finetune from the XDoG-style model on other sketch styles instead of training all styles together (ablation) | Contextual GAN · CelebA/CUB sketch styles | qualitative generation quality | Improved |
| add KL-divergence-based best-of-N initialization of the latent code (ablation) | Contextual GAN · test-time latent projection | convergence and failure rate (qualitative) | Improved |
| use contextual GAN on badly drawn hand sketches instead of pix2pix or CE | sketch-to-image generation · 50 free-hand sketches per category (face, bird, car) | visual realism (qualitative) | Improved |
Closest papers by meaning
- SketchGAN: Joint Sketch Completion and Recognition With Generative Adversarial Network (CVPR 2019 · similarity 0.92)
- SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis (CVPR 2018 · similarity 0.90)
- Sketch Your Own GAN (ICCV 2021 · similarity 0.89)
- SketchyCOCO: Image Generation From Freehand Scene Sketches (CVPR 2020 · similarity 0.89)
- Constrained Generative Adversarial Networks for Interactive Image Generation (CVPR 2019 · similarity 0.89)
- APDrawingGAN: Generating Artistic Portrait Drawings From Face Photos With Hierarchical GANs (CVPR 2019 · 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.
