Digital Human Driving via Multi-Angle Key Point Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital human driving methods using single-angle video capture result in visible shaking, malposition of joint rotation, and partial part loss due to undetected key points, and require costly calibration plates for multi-lens cameras.
Innovation Solution
Employ multiple video capture devices to capture video data from various angles, determining key point coordinates and mapping relationships to drive digital humans in a virtual 3D space, eliminating the need for calibration plates and improving movement efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a monocular camera is used to capture video data, then the device complexity is reduced, but the measurement precision of key points deteriorates due to blind areas
Solution Approach 1:
The patent divides the single viewing task into multiple segments by using multiple cameras positioned at different locations. Each camera captures key points from its own perspective, and the system segments the overall measurement task across multiple detection devices to eliminate blind spots and improve comprehensive measurement precision.
Solution Approach 2:
The patent merges data from multiple cameras by establishing a unified coordinate system and fusing key point information from different viewing angles. This combination allows the system to overcome the limitations of individual cameras and achieve complete coverage of the target object's key points.
2Measurement precision
If multiple video capture devices are used to capture from multiple angles, then the measurement precision of key points is improved, but the device complexity increases
Solution Approach 1:
The patent creates a universal coordinate system that can handle data from multiple cameras with different positions and orientations. This multi-functional coordinate framework allows the system to process measurements from various angles uniformly, reducing the complexity burden of having multiple devices.
Solution Approach 2:
The patent introduces a coordinate transformation mechanism as an intermediary that bridges multiple camera coordinate systems. This mediator converts local measurements from each camera into the unified coordinate system, simplifying the integration process and reducing overall system complexity.
3Measurement precision
If calibration plates are used for multi-lens cameras, then the measurement precision is improved, but the ease of manufacture deteriorates
Solution Approach 1:
The patent enables the system to perform self-calibration by using the target object itself as the calibration reference. Instead of requiring external calibration plates, the system captures images of the target from multiple angles and automatically computes transformation parameters, making the calibration process intrinsic to the measurement process itself.
Solution Approach 2:
The patent creates virtual copies of the target object in the unified coordinate system from multiple camera perspectives. By reconstructing the target's position and orientation in 3D space from 2D images, the system eliminates the need for physical calibration plates while maintaining measurement precision.
Data Source
AI summary
A digital human driving method and apparatus are provided which relate to computer and image processing, and can solve the problem of shaking, joint rotation malposition and partial loss of a digital human during a driving process. The solution includes: capturing video data from multiple angles of view in a real three-dimensional space by multiple video capture devices; determining a first coordinate of a key point of the target human; determining a mapping relationship based on the first coordinate; calculating a second coordinate based on the mapping relationship and the first coordinate; processing the second coordinate according to a key point rotation model to obtain rotation value of the virtual key point in the virtual three-dimensional space; and driving the digital human to move based on the rotation value of the virtual key point in the virtual three-dimensional space.