MangaNinja: Line Art Colorization with Precise Reference Following

TL;DR

MangaNinja introduces a diffusion-based method for reference-guided line art colorization, achieving precise character detail transfer and interactive control, surpassing existing solutions in accuracy and versatility.

Contribution

The paper presents novel patch shuffling and point-driven control mechanisms that enhance detail accuracy and user interaction in line art colorization.

Findings

Outperforms current methods in colorization precision

Effective in challenging and cross-character cases

Demonstrates flexible multi-reference harmonization

Abstract

Derived from diffusion models, MangaNinjia specializes in the task of reference-guided line art colorization. We incorporate two thoughtful designs to ensure precise character detail transcription, including a patch shuffling module to facilitate correspondence learning between the reference color image and the target line art, and a point-driven control scheme to enable fine-grained color matching. Experiments on a self-collected benchmark demonstrate the superiority of our model over current solutions in terms of precise colorization. We further showcase the potential of the proposed interactive point control in handling challenging cases, cross-character colorization, multi-reference harmonization, beyond the reach of existing algorithms.