video-similarity-search

Near Duplicate Video Detection (Perceptual Video Hashing) - Get a 64-bit comparable hash-value for any video.

Authors official PyTorch implementation of the "ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning" [ICCV 2019]

Authors official PyTorch implementation of the "DnS: Distill-and-Select for Efficient and Accurate Video Indexing and Retrieval" [IJCV 2022]

Authors official PyTorch implementation of the "Self-Supervised Video Similarity Learning" [CVPRW 2023]

Python library for finding similar content in videos.

一款快速、精确的视频相似度检测 分析工具+视频合并工具，扫描指定目录中的视频，通过动态抽帧、画面预处理、CLIP 特征、FAISS 检索和时间一致性分析，计算视频之间的双向覆盖率、整体相似度、匹配片段和时间窗口结果。

A video similarity pipeline extracting frame embeddings using OpenAI’s CLIP model and aggregating them into video-level representations. The system indexes embeddings with FAISS to enable efficient, real-time similarity search across varied video collections.

Video database querying platform