Research map / Image generation / Text-to-image diffusion
TiGAN: Text-Based Interactive Image Generation and Manipulation
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 TiGAN instead of XMC-GAN for text-to-image generation | MS-COCO 2014 text-to-image generation · MS-COCO 2014 | FID: 9.33 -> 8.90 | Helped |
| Use TiGAN instead of XMC-GAN for text-to-image generation | MS-COCO 2014 text-to-image generation · MS-COCO 2014 | IS: 30.45 -> 31.95 | Helped |
| Use TiGAN instead of fine-tuned TediGAN for text-to-image generation | Multi-modal CelebA-HQ · Multi-modal CelebA-HQ | FID: 27.39 -> 11.35 | Helped |
| Use TiGAN instead of fine-tuned TediGAN for text-to-image generation | Multi-modal CelebA-HQ · Multi-modal CelebA-HQ | IS: 2.29 -> 2.85 | Helped |
| Add generator text-image contrastive loss LCLIP (ablation) | TiGAN · MS-COCO 2014 | IS: 22.87 -> 31.95 | Helped |
| Add generator text-image contrastive loss LCLIP (ablation) | TiGAN · MS-COCO 2014 | FID: 19.62 -> 8.90 | Helped |
| Add discriminator contrastive loss LCD (ablation) | TiGAN · MS-COCO 2014 | IS: 27.21 -> 31.95 | Helped |
| Add discriminator contrastive loss LCD (ablation) | TiGAN · MS-COCO 2014 | FID: 18.21 -> 8.90 | Helped |
| Add threshold-based selection of relevant style dimensions during manipulation (ablation) | TiGAN · UT Zappos50K interactive | SR (10): 0.628 -> 0.682 | Helped |
| Add threshold-based selection of relevant style dimensions during manipulation (ablation) | TiGAN · Multi-modal CelebA-HQ interactive | SR (10): 0.737 -> 0.761 | Helped |
| Use TiGAN instead of SeqAttnGAN | UT Zappos50K interactive generation · UT Zappos50K interactive | AR (10): 7.090 -> 5.412 | Helped |
| Use TiGAN instead of SeqAttnGAN | Multi-modal CelebA-HQ interactive generation · Multi-modal CelebA-HQ | SR (10): 0.582 -> 0.761 | Helped |
Closest papers by meaning
- InstantPainting: Expanding GANs for Efficient Text-Conditioned Image Generation Platform (AAAI 2025 · similarity 0.87)
- Prompt Augmentation for Self-supervised Text-guided Image Manipulation (CVPR 2024 · similarity 0.87)
- Rich Human Feedback for Text-to-Image Generation (CVPR 2024 · similarity 0.86)
- E$^2$GAN: Efficient Training of Efficient GANs for Image-to-Image Translation (ICML 2024 · similarity 0.86)
- Focus-N-Fix: Region-Aware Fine-Tuning for Text-to-Image Generation (CVPR 2025 · similarity 0.86)
- Adversarial Text to Continuous Image Generation (CVPR 2024 · similarity 0.85)
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.