提升文本引导图像编辑的细节与速度，解决模糊和失效问题。
LGCC: Enhancing Flow Matching Based Text-Guided Image Editing with Local Gaussian Coupling and Context Consistency
- 用局部高斯耦合保留图像细节，避免噪声破坏结构。
- 引入内容一致性损失，确保修改符合指令，不删不该删的内容。
- 速度提升2-5倍，仅需原方法40%-50%时间，适合高效编辑场景。
基于流匹配的多模态大模型在图像编辑中展现出巨大潜力，但当前先进方法如BAGEL存在细节退化、内容不一致和效率低的问题，源于依赖随机噪声初始化。为此，我们提出LGCC框架，包含两个核心组件：局部高斯噪声耦合（LGNC）和内容一致性损失（CCL）。LGNC通过将目标图像嵌入与其局部扰动版本建模为耦合对，保持空间细节；CCL则确保编辑指令与图像修改间的语义对齐，防止意外内容移除。通过课程学习将LGCC集成至预训练的BAGEL模型，显著减少推理步数，在I2EBench上局部细节得分提升1.60%，整体得分提升0.53%。LGCC实现轻量编辑3-5倍加速，通用编辑2倍加速，仅需BAGEL或Flux 40%-50%的推理时间。结果表明，LGCC在保留细节、维持上下文一致性和提升推理速度方面表现优异，提供无需牺牲质量的高效解决方案。
原文摘要 · Abstract (English)
Recent advancements have demonstrated the great potential of flow matching-based Multimodal Large Language Models (MLLMs) in image editing. However, state-of-the-art works like BAGEL face limitations, including detail degradation, content inconsistency, and inefficiency due to their reliance on random noise initialization. To address these issues, we propose LGCC, a novel framework with two key components: Local Gaussian Noise Coupling (LGNC) and Content Consistency Loss (CCL). LGNC preserves spatial details by modeling target image embeddings and their locally perturbed counterparts as coupled pairs, while CCL ensures semantic alignment between edit instructions and image modifications, preventing unintended content removal. By integrating LGCC with the BAGEL pre-trained model via curriculum learning, we significantly reduce inference steps, improving local detail scores on I2EBench by 1.60% and overall scores by 0.53%. LGCC achieves 3x -- 5x speedup for lightweight editing and 2x for universal editing, requiring only 40% -- 50% of the inference time of BAGEL or Flux. These results demonstrate LGCC's ability to preserve detail, maintain contextual integrity, and enhance inference speed, offering a cost-efficient solution without compromising editing quality.
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。