Incremental completion method and device of knowledge graph, electronic equipment and storage medium
By extracting visual element information from video streams and generating individual behavioral events, the problems of large data processing volume and low accuracy in constructing graphs from video data are solved, achieving efficient and flexible graph completion effects.
CN122416342APending Publication Date: 2026-07-17
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-05-22
- Publication Date
- 2026-07-17
Smart Images
Figure CN122416342A_ABST
Abstract
本发明公开了一种知识图谱的增量补全方法、装置、电子设备及存储介质，其中，该方法包括：获取原始视频流，并从原始视频流中提取至少一个待检测对象的视觉要素信息；针对各待检测对象，对当前待检测对象的视觉要素信息进行语义识别，得到当前待检测对象对应的个体行为事件；基于个体行为事件，生成当前待检测对象对应的待融合子图；调取个体行为事件所属的目标事件类别对应的初始知识图谱，并基于待融合子图对初始知识图谱进行图谱补全操作，得到目标知识图谱。实现降低数据处理量，提高图谱完善效率和图谱完善的准确性，并实现灵活对不同事件类别的初始知识图谱进行图谱补全更新的效果。
Need to check novelty before this filing date? Find Prior Art