VideoSum
Extracts content from multiple video platforms and generates intelligent knowledge graphs with OCR support.
About
VideoSum extracts content from popular Chinese video platforms like Douyin, Bilibili, Xiaohongshu, and Zhihu. Leveraging OCR technology, it extracts text from images and generates intelligent knowledge graphs, providing structured content analysis and relationship mapping. The server utilizes context-aware extraction and anti-crawling strategies to ensure optimal results, making it a robust solution for content summarization and analysis from diverse online video sources.
Key Features
- Multi-platform support (Douyin, Bilibili, Xiaohongshu, Zhihu)
- 1 GitHub stars
- OCR text recognition using PaddleOCR
- Context-aware content extraction
- Chinese content optimization
- Knowledge graph generation for intelligent content structuring