Research map / Video / Video-language understanding
PeCA: Palette Context Assisted Inference for Test-Time Paint-Bucket Colourisation on Animation Videos
Dongheng Lin, Jianbo Jiao
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 Active Reference Expansion (ablation) | DACoN 1.1 · PBC-3D design-sheet one-shot | Acc (%): 68.01 -> 69.04 | Improved |
| add Probability Aggregation on top of ARE (ablation) | DACoN 1.1 · PBC-3D design-sheet one-shot | Acc (%): 69.04 -> 70.61 | Improved |
| add cyclic-gated temporal fusion on top of ARE and PA (ablation) | DACoN 1.1 · PBC-3D design-sheet one-shot | Acc (%): 70.61 -> 72.04 | Improved |
| add Active Reference Expansion (ablation) | DINOv2 ViT-L/14 · PBC-3D design-sheet one-shot | Acc (%): 57.49 -> 58.74 | Improved |
| add Probability Aggregation on top of ARE (ablation) | DINOv2 ViT-L/14 · PBC-3D design-sheet one-shot | Acc (%): 58.74 -> 60.64 | Improved |
| add ARE to PA and CT (ablation) | DINOv2 ViT-L/14 · PBC-3D design-sheet one-shot | Acc (%): 58.68 -> 61.38 | Improved |
| add ARE and temporal fusion without probability aggregation (ablation) | DINOv2 ViT-L/14 · PBC-3D design-sheet one-shot | Acc (%): 57.49 -> 58.51 | No clear effect |
| add PeCA to frozen SAM2.1-Large | SAM2.1-Large · PBC-3D design-sheet one-shot | Acc (%): 34.54 -> 46.65 | Improved |
| add PeCA to trained DACoN 1.1 | DACoN 1.1 · PBC-3D design-sheet one-shot | Acc (%): 68.01 -> 72.04 | Improved |
| add PeCA to DINOv2 with 5-shot references | DINOv2 ViT-L/14 · PBC-3D 5-shot refs | Acc (%): 62.65 -> 66.46 | Improved |
| add PeCA to DACoN 1.1 on long in-between video | DACoN 1.1 · Anita-Pirate in-between | Acc (%): 38.16 -> 41.24 | Improved |
| add PeCA to DACoN 1.1 on real hand-drawn first-frame colourisation | DACoN 1.1 · PBC-Real first-frame | Acc (%): 65.82 -> 67.64 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AniDoc — AniDoc: Animation Creation Made Easier (CVPR 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Image Referenced Sketch Colorization Based on Animation Creation Workflow (CVPR 2025 · similarity 0.87)
- Learning Inclusion Matching for Animation Paint Bucket Colorization (CVPR 2024 · similarity 0.86)
- VIRES: Video Instance Repainting via Sketch and Text Guided Generation (CVPR 2025 · similarity 0.84)
- Automatic Controllable Colorization via Imagination (CVPR 2024 · similarity 0.83)
- AniDoc: Animation Creation Made Easier (CVPR 2025 · similarity 0.83)
- MangaNinja: Line Art Colorization with Precise Reference Following (CVPR 2025 · similarity 0.83)
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.
