用语义先验提升大块缺失图像的修复质量
CSF-Net: Context-Semantic Fusion Network for Large Mask Inpainting
- 引入预训练模型生成结构感知候选图作为语义先验
- 融合上下文与语义信息，显著降低物体幻觉
- 可无缝接入现有模型，适用于多种遮挡场景
本文提出一种语义引导框架，解决大范围缺失区域图像修复难题。为弥补上下文信息不足，利用预训练的非可视补全（Amodal Completion, AC）模型生成结构感知候选图，作为缺失区域的语义先验。提出基于Transformer的上下文-语义融合网络（CSF-Net），将候选图与上下文特征融合，生成用于修复的语义引导图，提升结构准确性和语义一致性。该方法无需修改现有修复模型架构，可在多种遮挡条件下持续提升性能。在Places365和COCOA数据集上的大量实验表明，CSF-Net有效减少物体幻觉，增强视觉真实感与语义对齐度。
原文摘要 · Abstract (English)
In this paper, we propose a semantic-guided framework to address the challenging problem of large-mask image inpainting, where essential visual content is missing and contextual cues are limited. To compensate for the limited context, we leverage a pretrained Amodal Completion (AC) model to generate structure-aware candidates that serve as semantic priors for the missing regions. We introduce Context-Semantic Fusion Network (CSF-Net), a transformer-based fusion framework that fuses these candidates with contextual features to produce a semantic guidance image for image inpainting. This guidance improves inpainting quality by promoting structural accuracy and semantic consistency. CSF-Net can be seamlessly integrated into existing inpainting models without architectural changes and consistently enhances performance across diverse masking conditions. Extensive experiments on the Places365 and COCOA datasets demonstrate that CSF-Net effectively reduces object hallucination while enhancing visual realism and semantic alignment. The code for CSF-Net is available at https://github.com/chaeyeonheo/CSF-Net.
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。