通过追踪像素位置提升图像复制检测的精度与可解释性。
Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection
- 引入像素坐标追踪模块，显式保持编辑过程中的空间映射。
- 提出基于重叠比的对比损失，增强局部特征匹配准确性。
- 在DISC21数据集上达到88.7% uAP，优于现有方法且结果更易解释。
图像复制检测（ICD）旨在通过鲁棒的特征表示学习识别图像对之间的篡改内容。尽管自监督学习（SSL）已推动ICD系统发展，但现有视图级对比方法因细粒度对应关系学习不足，在复杂编辑下表现受限。本文通过利用编辑内容中的固有几何可追溯性，提出两项关键创新：首先，设计PixTrace像素坐标追踪模块，跨编辑变换维持显式空间映射；其次，提出CopyNCE几何引导对比损失，基于PixTrace验证映射计算的重叠比正则化局部块相似性。该方法将像素级可追溯性与块级相似性学习相融合，抑制了SSL训练中的监督噪声。大量实验表明，本方法不仅在DISC21数据集上取得最佳性能（匹配器88.7% uAP / 83.9% RP90，描述符72.6% uAP / 68.4% RP90），且相比现有方法具备更强可解释性。
原文摘要 · Abstract (English)
Image Copy Detection (ICD) aims to identify manipulated content between image pairs through robust feature representation learning. While self-supervised learning (SSL) has advanced ICD systems, existing view-level contrastive methods struggle with sophisticated edits due to insufficient fine-grained correspondence learning. We address this limitation by exploiting the inherent geometric traceability in edited content through two key innovations. First, we propose PixTrace - a pixel coordinate tracking module that maintains explicit spatial mappings across editing transformations. Second, we introduce CopyNCE, a geometrically-guided contrastive loss that regularizes patch affinity using overlap ratios derived from PixTrace's verified mappings. Our method bridges pixel-level traceability with patch-level similarity learning, suppressing supervision noise in SSL training. Extensive experiments demonstrate not only state-of-the-art performance (88.7% uAP / 83.9% RP90 for matcher, 72.6% uAP / 68.4% RP90 for descriptor on DISC21 dataset) but also better interpretability over existing methods.
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。