# DACoN: DINO for Anime Paint Bucket Colorization with Any Number of Reference Images

arXiv: https://arxiv.org/abs/2509.14685 · HTML: https://mlsanity.com/paper/2509.14685/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 50 references that are in the map.
- First posted on arXiv 18 Sep 2025.

## Builds on (most cited first, 26 of 26)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [SuperGlue: Learning Feature Matching with Graph Neural Networks](https://mlsanity.com/paper/1911.11763/) (2019)
- [Emergent Correspondence from Image Diffusion](https://mlsanity.com/paper/2306.03881/) (2023)
- [A Tale of Two Features: Stable Diffusion Complements DINO for Zero-Shot Semantic Correspondence](https://mlsanity.com/paper/2305.15347/) (2023)
- [ToonCrafter: Generative Cartoon Interpolation](https://mlsanity.com/paper/2405.17933/) (2024)
- [Deep Animation Video Interpolation in the Wild](https://mlsanity.com/paper/2104.02495/) (2021)
- [Reference-Based Sketch Image Colorization using Augmented-Self Reference and Dense Semantic Correspondence](https://mlsanity.com/paper/2005.05207/) (2020)
- [LVCD: Reference-based Lineart Video Colorization with Diffusion Models](https://mlsanity.com/paper/2409.12960/) (2024)
- [Eliminating Gradient Conflict in Reference-based Line-Art Colorization](https://mlsanity.com/paper/2207.06095/) (2022)
- [AniDoc: Animation Creation Made Easier](https://mlsanity.com/paper/2412.14173/) (2024)
- [The Animation Transformer: Visual Correspondence via Segment Matching](https://mlsanity.com/paper/2109.02614/) (2021)
- [Comicolorization: Semi-Automatic Manga Colorization](https://mlsanity.com/paper/1706.06759/) (2017)
- [Generative AI for Cel-Animation: A Survey](https://mlsanity.com/paper/2501.06250/) (2025)
- [MangaNinja: Line Art Colorization with Precise Reference Following](https://mlsanity.com/paper/2501.08332/) (2025)
- [Learning Inclusion Matching for Animation Paint Bucket Colorization](https://mlsanity.com/paper/2403.18342/) (2024)
- [AnimeRun: 2D Animation Visual Correspondence from Open Source 3D Movies](https://mlsanity.com/paper/2211.05709/) (2022)
- [SketchBetween: Video-to-Video Synthesis for Sprite Animation via Sketches](https://mlsanity.com/paper/2209.00185/) (2022)
- [ColorFlow: Retrieval-Augmented Image Sequence Colorization](https://mlsanity.com/paper/2412.11815/) (2024)
- [Deep Line Art Video Colorization with a Few References](https://mlsanity.com/paper/2003.10685/) (2020)
- [Efficient semantic image segmentation with superpixel pooling](https://mlsanity.com/paper/1806.02705/) (2018)
- [Paint Bucket Colorization Using Anime Character Color Design Sheets](https://mlsanity.com/paper/2410.19424/) (2024)

## Built on by (most cited first, 2 of 2)

- [See-through: Single-image Layer Decomposition for Anime Characters](https://mlsanity.com/paper/2602.03749/) (2026)
- [PeCA: Palette Context Assisted Inference for Test-Time Paint-Bucket Colourisation on Animation Videos](https://mlsanity.com/paper/2608.00903/) (2026)

## Abstract (arXiv)

Automatic colorization of line drawings has been widely studied to reduce the labor cost of hand-drawn anime production. Deep learning approaches, including image/video generation and feature-based correspondence, have improved accuracy but struggle with occlusions, pose variations, and viewpoint changes. To address these challenges, we propose DACoN, a framework that leverages foundation models to capture part-level semantics, even in line drawings. Our method fuses low-resolution semantic features from foundation models with high-resolution spatial features from CNNs for fine-grained yet robust feature extraction. In contrast to previous methods that rely on the Multiplex Transformer and support only one or two reference images, DACoN removes this constraint, allowing any number of references. Quantitative and qualitative evaluations demonstrate the benefits of using multiple reference images, achieving superior colorization performance. Our code and model are available at https://github.com/kzmngt/DACoN.