基于人工智能的人形相关数据可视化方法及系统
By using high-dimensional probabilistic keypoint processing, graph convolution, inverse kinematics, and self-attention mechanism completion, combined with spatiotemporal graph convolution and conditional random fields, the problems of joint proportion imbalance and spatial position distortion in human-related data visualization are solved, achieving high-precision human recognition and risk assessment, and generating highly diverse 3D situation maps.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHAANXI JIANXIE BIOTECHNOLOGY GROUP CO LTD
- Filing Date
- 2025-12-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies are prone to joint proportion imbalance and spatial position distortion in human-related data visualization due to missing depth information and projection ambiguity. Single-frame estimation methods are sensitive to temporal jitter and occlusion, resulting in unstable output and missing joints.
We employ high-dimensional probabilistic key point processing, graph convolution, inverse kinematics Kalman filtering, and self-attention mechanism completion, combined with spatiotemporal graph convolution and conditional random fields, to utilize BIM models for 3D situation map rendering and interactive safety monitoring interface generation.
It significantly improves the robustness and accuracy of human figure recognition, enhances the accuracy of behavior recognition and the real-time performance of risk assessment in complex scenarios, and generates diverse and comparative 3D situation maps.
Smart Images
Figure CN121746593B_ABST