一种自动化拍摄人形机器人及其控制方法
By constructing a target composition constraint model and a multimodal perception system, we have achieved human-like interaction and intelligent composition of automated shooting equipment, which solves the problems of insufficient intelligent composition and fragmented motion control in existing equipment, and improves shooting quality and application breadth.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHENGDU WUJING YUNKE TECHNOLOGY CO LTD
- Filing Date
- 2026-06-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing automated shooting equipment lacks intelligent composition, has a disconnect between motion control and shooting logic, and lacks human-like interactive capabilities, resulting in images that lack narrative and visual appeal. Furthermore, in scenes with multiple subjects, it can easily interfere with eye contact or cause obstruction.
By constructing a target composition constraint model, combining a multimodal perception system and a driving system, the robot's shooting pose is calculated in real time. A gaze contact simulation and interactive fine-tuning mechanism are introduced to optimize the optical axis pose of the camera components. Furthermore, avoidance constraints are introduced in multi-object shooting scenarios to achieve human-like shooting control.
It improves the image quality and narrative expressiveness of automated shooting, eliminates the mechanical feeling and discomfort, expands the scope of application, ensures human-machine safety and personalized adaptation, and enhances the system's reliability and emergency response capabilities.
Smart Images
Figure CN122401449A_ABST