D$^2$LV: A Data-Driven and Local-Verification Approach for Image Copy Detection
This addresses the problem of detecting copied images in social media, with a competitive but incremental improvement in a specific benchmark.
The paper tackled image copy detection by proposing a data-driven and local-verification approach that uses unsupervised pre-training and a global-local matching strategy, achieving first place out of 1,103 participants in the Facebook AI Image Similarity Challenge.
Image copy detection is of great importance in real-life social media. In this paper, a data-driven and local-verification (D$^2$LV) approach is proposed to compete for Image Similarity Challenge: Matching Track at NeurIPS'21. In D$^2$LV, unsupervised pre-training substitutes the commonly-used supervised one. When training, we design a set of basic and six advanced transformations, and a simple but effective baseline learns robust representation. During testing, a global-local and local-global matching strategy is proposed. The strategy performs local-verification between reference and query images. Experiments demonstrate that the proposed method is effective. The proposed approach ranks first out of 1,103 participants on the Facebook AI Image Similarity Challenge: Matching Track. The code and trained models are available at https://github.com/WangWenhao0716/ISC-Track1-Submission.