无需训练，实时优化漫画编辑轨迹，保留整体构图
Inference-time Trajectory Optimization for Structure-Preserving Manga Image Editing
- 用输入图像自适应修正编辑路径，不需额外训练
- 文本移除任务表现提升，画面构图更稳定
- 轻量高效，推理耗时仅增加11%，适合实际应用
我们提出一种轻量级、无需训练的轨迹修正方法，仅依赖输入漫画图像即可适配预训练编辑模型。尽管预训练图像编辑模型进展迅速，但其主要在自然图像上训练，对漫画效果不佳；而针对漫画重新训练或微调成本高且存在版权风险。实践中多数漫画编辑任务要求结构保持，即局部细节修改的同时保留全局构图。为此，本方法通过锚定空提示重建轨迹来修正早期编辑路径。实验表明，在主要文本移除任务中性能提升，定性结果显示屏线合成时构图保留更佳。在RTX A6000上使用FLUX.1 Kontext时，运行时间增加11%，峰值内存仅增0.1%；在NVIDIA H200上使用Qwen Image Edit 2509时，运行时间增加仅为0.1%。
原文摘要 · Abstract (English)
We present a lightweight, training-free trajectory correction method that adapts a pretrained image editing model to each input manga image using only the input itself. Despite recent progress in pretrained image editing, such models often underperform on manga because they are trained predominantly on natural-image data, while re-training or fine-tuning them on manga is costly and raises copyright concerns. Many manga image editing tasks encountered in practice are structure-preserving, requiring local details to be modified while the input's global composition is retained. To support this common editing setting, our method corrects the early editing trajectory by anchoring it to an empty-prompt reconstruction trajectory. Experiments indicate improved performance in the main text-removal setting, while qualitative examples suggest better composition preservation in screentone synthesis. With FLUX.1 Kontext on an RTX A6000, the method incurs 11% runtime overhead and 0.1% peak-memory overhead; an additional runtime measurement with Qwen Image Edit 2509 on an NVIDIA H200 shows only a 0.1% increase.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。