MangaNinja: Line Art Colorization with Precise Reference Following
This work addresses the specific problem of accurate colorization for manga line art, which is incremental in improving reference-following techniques.
The paper tackles the problem of reference-guided line art colorization by developing MangaNinja, a diffusion-based model that incorporates patch shuffling and point-driven control for precise color matching, achieving superior performance on a self-collected benchmark over existing solutions.
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.