Research map / Vision-language and multimodal / Cross-modal learning
OmniColor: A Unified Framework for Multi-modal Lineart Colorization
Xulu Zhang, Haoqian DU, Xiao-Yong Wei, Li Qing
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 VLM-only semantic encoding (ablation) | OmniColor backbone · Ablation (L+T+H) | Inference time (s): 452 -> 203 | Improved |
| add VLM-only semantic encoding (ablation) | OmniColor backbone · Ablation (L+T+H) | FID: 110.32 -> 112.29 | No clear effect |
| add DFA loss (ablation) | OmniColor · Ablation (L+T+H) | SSIM: 0.7287 -> 0.7500 | Improved |
| add DFA loss (ablation) | OmniColor · Ablation (L+T+H) | FID: 112.29 -> 105.90 | Improved |
| add DFA loss (ablation) | OmniColor · Ablation (L+T+H) | Consistency error: 0.61 -> 0.17 | Improved |
| add TRE module (ablation) | OmniColor · Ablation (L+T+H) | FID: 105.90 -> 97.31 | Improved |
| add TRE module (ablation) | OmniColor · Ablation (L+T+H) | History tokens: 7521 -> 5203 | Improved |
| add TRE module (ablation) | OmniColor · Ablation (L+T+H) | Inference time (s): 203 -> 150 | Improved |
| add TRE module (ablation) | OmniColor · Ablation (L+T+H) | Consistency error: 0.17 -> 0.56 | Hurt |
| add AS-Gate (ablation) | OmniColor · Ablation (L+T+H) | FID: 97.31 -> 91.49 | Improved |
| add AS-Gate (ablation) | OmniColor · Ablation (L+T+H) | Consistency error: 0.56 -> 0.37 | Improved |
| add sequential generation (H) to L+T | OmniColor · Lineart colorization test set | Consistency error: 5.44 -> 0.37 | Improved |
Closest papers by meaning
- Control Color: Multimodal Diffusion-Based Interactive Image Colorization (IJCV 2025 · similarity 0.88)
- Recolour What Matters: Region-Aware Colour Editing via Token-Level Diffusion (ECCV 2026 · similarity 0.88)
- Structural Consistency and Controllability for Diverse Colorization (ECCV 2018 · similarity 0.88)
- Training-Free Text-Guided Color Editing with Multi-Modal Diffusion Transformer (ICLR 2026 · similarity 0.87)
- Automatic Controllable Colorization via Imagination (CVPR 2024 · similarity 0.86)
- MangaNinja: Line Art Colorization with Precise Reference Following (CVPR 2025 · 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.
