Depth Image User Behavior Extraction for Floor-Based Positions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional skeletal tracking techniques are unreliable for detecting user behaviors such as laying or sitting on or near the floor, limiting their effectiveness in applications like gaming and multimedia interactions.
Innovation Solution
The use of depth images to extract user behavior by fitting curves to pixels corresponding to the user, determining angles and curvatures of the body, and identifying average extremity positions, which can supplement or replace skeletal tracking methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If skeletal tracking techniques are used to detect user behavior, then detection capability is provided for certain types of user behaviors, but reliability deteriorates for user behaviors where the user is laying or sitting on or near the floor
Solution Approach 1:
The patent transitions from 2D skeletal tracking to 3D depth image analysis. By capturing depth information and creating a three-dimensional representation of the user's body, the system can reliably detect user behaviors in any position including laying or sitting on the floor, which were unreliable in traditional 2D skeletal tracking.
Solution Approach 2:
The patent changes the fundamental parameters used for detection from skeletal joint coordinates in 2D space to depth values and spatial coordinates in 3D space. This parameter transformation enables accurate detection of body angles, curvatures, and extremity positions regardless of the user's orientation or position relative to the camera.
2Ease of operation
If conventional controls (controllers, remotes, keyboards, mice) are used, then control functionality is provided, but ease of operation deteriorates due to difficulty to learn and barrier between user and application
Solution Approach 1:
The system uses the user's own body movements as the control interface. The user's natural motions directly control the application without requiring learning of abstract control schemes. The depth imaging system captures the user's movements and translates them directly into application actions, making the interface intuitive and self-explanatory.
Solution Approach 2:
The patent replaces mechanical control devices (controllers, keyboards, mice) with an optical sensing system. Instead of requiring physical interaction with mechanical controls, the system uses depth cameras to capture and interpret the user's body movements, substituting mechanical input methods with optical field-based detection.
3Measurement precision
If skeletal tracking techniques are used, then detection capability is provided, but measurement precision deteriorates for body angles and curvatures when users are in positions where skeletal tracking fails
Solution Approach 1:
By moving to 3D depth space, the system can accurately measure body angles and curvatures from multiple spatial dimensions. The depth information provides additional measurement axes that enable precise calculation of anatomical angles and body segment curvatures regardless of the user's position relative to the camera.
Solution Approach 2:
The system changes from detecting discrete skeletal joint positions to analyzing continuous depth profiles across the entire body. This parameter transformation enables precise measurement of body angles and curvatures by fitting curves to depth image data, providing accurate measurements even when the user is laying or sitting on the floor.
Data Source
AI summary
Embodiments described herein use depth images to extract user behavior, wherein each depth image specifies that a plurality of pixels correspond to a user. In certain embodiments, one or more average extremity positions of a user, which can also be referred to as average positions of extremity blobs, are extracted from a depth image. An application is then updated based on the average positions of extremity blobs.