A Similarity Alignment Model for Video Copy Segment Matching

TL;DR

This paper presents a novel Similarity Alignment Model (SAM) for video copy segment matching, achieving top performance in the CVPR 2023 Video Similarity Challenge by significantly outperforming competitors.

Contribution

The paper introduces the SAM, a new model architecture specifically designed for video copy segment matching, demonstrating superior accuracy in a competitive benchmark.

Findings

SAM outperforms competitors with a 0.108 / 0.144 absolute improvement.

The model achieves state-of-the-art results in the CVPR 2023 Video Similarity Challenge.

Code implementation is publicly available for reproducibility.

Abstract

With the development of multimedia technology, Video Copy Detection has been a crucial problem for social media platforms. Meta AI hold Video Similarity Challenge on CVPR 2023 to push the technology forward. In this report, we share our winner solutions on Matching Track. We propose a Similarity Alignment Model(SAM) for video copy segment matching. Our SAM exhibits superior performance compared to other competitors, with a 0.108 / 0.144 absolute improvement over the second-place competitor in Phase 1 / Phase 2. Code is available at https://github.com/FeipengMa6/VSC22-Submission/tree/main/VSC22-Matching-Track-1st.