Google llc (20240202969). Depth-Based 3D Human Pose Detection and Tracking simplified abstract
Contents
Depth-Based 3D Human Pose Detection and Tracking
Organization Name
Inventor(s)
Depth-Based 3D Human Pose Detection and Tracking - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240202969 titled 'Depth-Based 3D Human Pose Detection and Tracking
The abstract of this patent application describes a method that involves determining the initial three-dimensional position of each keypoint representing predetermined body locations of an actor, receiving image and depth data of the actor's body, calculating visibility and depth field values for each keypoint, and updating the 3D position based on a loss value.
- The method determines the initial 3D position of keypoints representing body locations.
- It utilizes image and depth data to calculate visibility and depth field values for each keypoint.
- Based on these values, a loss value is determined for each keypoint.
- The 3D position of each keypoint is updated using the loss value.
Potential Applications: - Motion capture for animation and virtual reality applications - Biomechanical analysis in sports science and physical therapy - Human-computer interaction for gesture recognition and control systems
Problems Solved: - Accurate tracking and positioning of body keypoints in three-dimensional space - Improving the realism and precision of motion capture systems - Enhancing the efficiency of biomechanical analysis and human-computer interaction
Benefits: - Enhanced realism and accuracy in animation and virtual reality - Improved performance in sports science and physical therapy applications - More intuitive and responsive gesture-based control systems
Commercial Applications: Title: Three-Dimensional Keypoint Tracking System for Motion Capture and Biomechanical Analysis This technology can be used in industries such as entertainment (animation, gaming), sports science, healthcare (physical therapy), and human-computer interaction (gesture control systems). The market implications include improved product offerings, increased user engagement, and enhanced research capabilities.
Questions about Three-Dimensional Keypoint Tracking System: 1. How does this technology improve the accuracy of motion capture systems? This technology enhances accuracy by calculating visibility and depth field values for each keypoint, leading to more precise 3D positioning.
2. What are the potential applications of this system beyond animation and virtual reality? This system can also be used in sports science for biomechanical analysis and in healthcare for physical therapy applications.
Original Abstract Submitted
a method includes determining, for each respective keypoint of a plurality of keypoints that represents a plurality of predetermined body locations of an actor, an initial three-dimensional (3d) position of the respective keypoint. the method also includes receiving image data and depth data representing the body of the actor, and determining, for each respective keypoint, a visibility value based on a visibility of the respective keypoint in the image data and a depth field value based on the initial 3d position and a reference 3d position that is based on at least one nearest neighbor of the initial 3d position in the depth data. the method further includes determining, based on the visibility value and the depth field value of each respective keypoint, a loss value, and determining, for each respective keypoint, an updated 3d position of the respective keypoint based on the loss value.
