Inference-time Trajectory Optimization for Manga Image Editing
This work addresses the domain gap between natural images and manga for image editing, offering a practical solution that avoids retraining or fine-tuning large models.
The authors propose an inference-time trajectory optimization method that adapts a pretrained image editing model to manga images using only the input image, achieving consistent improvements over baselines with negligible computational overhead.
We present an inference-time adaptation method that tailors a pretrained image editing model to each input manga image using only the input image itself. Despite recent progress in pretrained image editing, such models often underperform on manga because they are trained predominantly on natural-image data. Re-training or fine-tuning large-scale models on manga is, however, generally impractical due to both computational cost and copyright constraints. To address this issue, our method slightly corrects the generation trajectory at inference time so that the input image can be reconstructed more faithfully under an empty prompt. Experimental results show that our method consistently outperforms existing baselines while incurring only negligible computational overhead.