用于标记图像之间的空间关系的强化学习模型
By training a machine learning model based on reinforcement learning, updating and optimizing the label mapping, the problem of high computational cost in image spatial relationship estimation in existing technologies is solved, and more efficient and accurate image correspondence search is achieved.
CN115362478BActive Publication Date: 2026-07-17OMRON CORP
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- OMRON CORP
- Filing Date
- 2021-03-26
- Publication Date
- 2026-07-17
Smart Images

Figure CN115362478B_ABST
Abstract
提供了一种计算系统,包括被配置为接收第一图像的标记映射的处理器。标记映射可以指示在包括在第一图像中的第一感兴趣区域和包括在第二图像中的第二感兴趣区域之间的空间关系。在经训练的强化学习模型中,处理器可以进一步被配置为至少基于标记映射、第一图像和第二图像来生成第一图像的经更新的标记映射。
Need to check novelty before this filing date? Find Prior Art