Robot Motion Recognition Using Virtual Joint Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion recognition processes for robots require significant computing resources and processing time due to the large amount of information needed to be processed, making them inefficient for real-time applications.
Innovation Solution
A method is introduced that generates a virtual joint image using joint location information and prediction values, allowing for more efficient motion recognition by dividing coordinate values into channels and applying a joint generation learning model to determine and arrange joint data based on body parts, enabling efficient processing and recognition of human motion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the entire image is used for motion recognition, then the details of motion recognition are improved, but the computing resources and processing time increase
Solution Approach 1:
The patent segments the image processing task by extracting only joint location information from the entire image. Instead of processing all image pixels for motion recognition, the system identifies and processes only the coordinates of key body joints, significantly reducing the data volume while maintaining motion recognition accuracy.
Solution Approach 2:
The patent extracts essential joint location data from the complete image. By using a joint detection model to identify and extract only the relevant joint coordinates (such as wrist, elbow, shoulder positions) from the full image, the system removes unnecessary image data while preserving the critical information needed for motion recognition.
2Productivity
If joint location information is extracted to reduce processing data, then computing resources are reduced, but motion recognition accuracy may deteriorate
Solution Approach 1:
The patent transforms the image data into a different parameter representation - specifically, converting visual image data into structured joint coordinate data (x, y positions of body joints). This parameter transformation maintains the essential motion information while reducing data complexity, enabling efficient processing without sacrificing recognition accuracy.
Data Source
AI summary
Disclosed is a computing device, which includes memory configured to store an image composed of a plurality of a plurality of frames where a person is captured as a subject, and a processor operatively connected with the memory. The processor may be configured to: determine a plurality of joints corresponding to the image; determine joint data; generate a virtual joint image comprising coordinate values; and store the generated virtual joint image in the memory. The joint data may include: joint prediction values for predicting whether the plurality of joints correspond to any of a plurality of known joints of the person, and joint location values corresponding to locations of the plurality of joints corresponding to the joint prediction values.