Depth Map Pose Estimation via Pixel Probability Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current tracking and processing systems face challenges in accurately interpreting user gestures and poses without knowing the body pose, leading to misinterpretation or missed commands, especially when users are at varying angles to the capture device.
Innovation Solution
The system processes depth map images from a depth camera to calculate the probability of each pixel being associated with body parts, constructs a model of the body by defining joints or nodes, and refines probabilities based on adjacent pixel associations, enabling accurate body pose estimation and gesture interpretation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system processes depth map images to calculate pixel probabilities and construct body models, then body pose estimation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent segments the body into multiple body parts (head, torso, limbs) and processes each part separately using probabilistic pixel association. This segmentation allows the complex problem of full-body pose estimation to be broken down into manageable parts, improving accuracy while controlling computational complexity through modular processing.
Solution Approach 2:
The patent transitions from 2D image processing to 3D pose estimation by utilizing depth map information. By incorporating the depth dimension, the system can accurately determine body pose in three-dimensional space, resolving the contradiction between measurement precision and complexity through the use of additional spatial information.
2Measurement precision
If the system refines probabilities based on adjacent pixel associations, then measurement precision is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary probability calculation for each pixel based on depth map information before refining through adjacent pixel associations. This preliminary action establishes a baseline that guides subsequent refinement steps, reducing the overall processing time by avoiding iterative optimization from scratch.
Solution Approach 2:
The system uses feedback from adjacent pixel associations to refine probability estimates. By incorporating local contextual information from neighboring pixels, the system improves measurement precision through iterative refinement while controlling processing time through efficient feedback loops.
Data Source
AI summary
Techniques for human body pose estimation are disclosed herein. Depth map images from a depth camera may be processed to calculate a probability that each pixel of the depth map is associated with one or more segments or body parts of a body. Body parts may then be constructed of the pixels and processed to define joints or nodes of those body parts. The nodes or joints may be provided to a system which may construct a model of the body from the various nodes or joints.