Research map / Image generation / Text-to-image diffusion
Coloring with Words: Guiding Image Colorization Through Text-based Palette Generation
Hyojin Bahng, Seungjoo Yoo, Wonwoong Cho, David Keetae Park, Ziming Wu, Xiaojuan Ma, Jaegul Choo
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 |
|---|---|---|---|
| add conditioning augmentation (CA) to the text encoder (ablation) | TPN · PAT test set | multimodality: 0.0 -> 5.43 | Improved |
| add conditioning augmentation (CA) to the text encoder (ablation) | TPN · PAT test set | palette diversity: 19.36 -> 20.82 | Improved |
| add the Huber loss to the cGAN objective (ablation) | TPN · User study Part I | fooling rate (%): 30.7 -> 56.2 | Improved |
| add the Huber loss to the cGAN objective (ablation) | TPN · PAT test set | KL divergence to training color distribution: 0.5833 -> 0.2299 | Improved |
| add the cGAN adversarial loss to Huber-only training (ablation) | TPN · PAT test set | KL divergence to training color distribution: 0.2375 -> 0.2299 | Improved |
| use TPN with CA instead of Heer and Stone palette model | palette generation · User study Part I | fooling rate (%): 39.6 -> 56.2 | Improved |
| use TPN with CA instead of Heer and Stone palette model | palette generation · PAT test set | palette diversity: 35.92 -> 20.82 | Hurt |
| use TPN with CA instead of Heer and Stone palette model | palette generation · PAT test set | multimodality: 0.0 -> 5.43 | Improved |
Closest papers by meaning
- PalGAN: Image Colorization with Palette Generative Adversarial Networks (ECCV 2022 · similarity 0.90)
- Colorful Image Colorization (ECCV 2016 · similarity 0.87)
- GenColorBench: A Color Evaluation Benchmark for Text-to-Image Generation (CVPR 2026 · similarity 0.87)
- DDColor: Towards Photo-Realistic Image Colorization via Dual Decoders (ICCV 2023 · similarity 0.87)
- Towards Vivid and Diverse Image Colorization With Generative Color Prior (ICCV 2021 · similarity 0.87)
- Learning Diverse Image Colorization (CVPR 2017 · similarity 0.87)
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.
