# Cobra: Efficient Line Art COlorization with BRoAder References

arXiv: https://arxiv.org/abs/2504.12240 · HTML: https://mlsanity.com/paper/2504.12240/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 53 references that are in the map.
- First posted on arXiv 16 Apr 2025.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [Efficiently Scaling Transformer Inference](https://mlsanity.com/paper/2211.05102/) (2022)
- [InstantID: Zero-shot Identity-Preserving Generation in Seconds](https://mlsanity.com/paper/2401.07519/) (2024)
- [Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Models](https://mlsanity.com/paper/2302.12228/) (2023)
- [BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion](https://mlsanity.com/paper/2403.06976/) (2024)
- [A Task is Worth One Word: Learning with Task Prompts for High-Quality Versatile Image Inpainting](https://mlsanity.com/paper/2312.03594/) (2023)
- [VideoBooth: Diffusion-based Video Generation with Image Prompts](https://mlsanity.com/paper/2312.00777/) (2023)
- [HumanSD: A Native Skeleton-Guided Diffusion Model for Human Image Generation](https://mlsanity.com/paper/2304.04269/) (2023)
- [TextureGAN: Controlling Deep Image Synthesis with Texture Patches](https://mlsanity.com/paper/1706.02823/) (2017)
- [Tag2Pix: Line Art Colorization Using Text Tag With SECat and Changing Loss](https://mlsanity.com/paper/1908.05840/) (2019)
- [Control Color: Multimodal Diffusion-based Interactive Image Colorization](https://mlsanity.com/paper/2402.10855/) (2024)
- [User-Guided Deep Anime Line Art Colorization with Conditional Adversarial Networks](https://mlsanity.com/paper/1808.03240/) (2018)
- [Diffusing Colors: Image Colorization with Text Guided Diffusion](https://mlsanity.com/paper/2312.04145/) (2023)
- [Comicolorization: Semi-Automatic Manga Colorization](https://mlsanity.com/paper/1706.06759/) (2017)
- [ColorFlow: Retrieval-Augmented Image Sequence Colorization](https://mlsanity.com/paper/2412.11815/) (2024)

## Built on by (most cited first, 6 of 6)

- [ToonComposer: Streamlining Cartoon Production with Generative Post-Keyframing](https://mlsanity.com/paper/2508.10881/) (2025)
- [SketchColour: Channel Concat Guided DiT-based Sketch-to-Colour Pipeline for 2D Animation](https://mlsanity.com/paper/2507.01586/) (2025)
- [TimeColor: Flexible Reference Colorization via Temporal Concatenation](https://mlsanity.com/paper/2601.00296/) (2026)
- [Uni-Animator: Towards Unified Visual Colorization](https://mlsanity.com/paper/2602.23191/) (2026)
- [Towards High-resolution and Disentangled Reference-based Sketch Colorization](https://mlsanity.com/paper/2603.05971/) (2026)
- [PeCA: Palette Context Assisted Inference for Test-Time Paint-Bucket Colourisation on Animation Videos](https://mlsanity.com/paper/2608.00903/) (2026)

## Abstract (arXiv)

The comic production industry requires reference-based line art colorization with high accuracy, efficiency, contextual consistency, and flexible control. A comic page often involves diverse characters, objects, and backgrounds, which complicates the coloring process. Despite advancements in diffusion models for image generation, their application in line art colorization remains limited, facing challenges related to handling extensive reference images, time-consuming inference, and flexible control. We investigate the necessity of extensive contextual image guidance on the quality of line art colorization. To address these challenges, we introduce Cobra, an efficient and versatile method that supports color hints and utilizes over 200 reference images while maintaining low latency. Central to Cobra is a Causal Sparse DiT architecture, which leverages specially designed positional encodings, causal sparse attention, and Key-Value Cache to effectively manage long-context references and ensure color identity consistency. Results demonstrate that Cobra achieves accurate line art colorization through extensive contextual reference, significantly enhancing inference speed and interactivity, thereby meeting critical industrial demands. We release our codes and models on our project page: https://zhuang2002.github.io/Cobra/.