![[2602.17484] Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.17484] Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection
Summary
The paper presents advancements in Image Copy Detection (ICD) by introducing PixTrace and CopyNCE, enhancing feature representation and interpretability in identifying manipulated images.
Why It Matters
As digital content manipulation becomes more prevalent, effective detection methods are crucial for maintaining content integrity. This research addresses limitations in current systems, providing a more robust approach to identifying edited images, which is essential for applications in security, copyright enforcement, and media verification.
Key Takeaways
- Introduces PixTrace for pixel coordinate tracking across edits.
- Presents CopyNCE, a contrastive loss that enhances patch affinity learning.
- Achieves state-of-the-art performance on the DISC21 dataset.
- Improves interpretability of results compared to existing methods.
- Addresses challenges in fine-grained correspondence learning.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.17484 (cs) [Submitted on 19 Feb 2026] Title:Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection Authors:Yichen Lu, Siwei Nie, Minlong Lu, Xudong Yang, Xiaobo Zhang, Peng Zhang View a PDF of the paper titled Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection, by Yichen Lu and 5 other authors View PDF HTML (experimental) Abstract: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 als...