Automatic Semantic Place Labeling via Activity Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic mapping and localization technologies require manual user input for labeling places, which is inefficient and labor-intensive, especially for distinguishing between similar indoor or outdoor locations.
Innovation Solution
An automated method using electronic device sensor data to recognize activities and determine semantic place labels by learning mappings between activities and locations, allowing for the automatic assignment of semantic labels to unlabeled places based on observed and typical mappings from multiple devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual user input is used for labeling places, then labeling accuracy can be ensured, but the process becomes labor-intensive and inefficient
Solution Approach 1:
The system performs automatic semantic labeling by having the place labeling system itself gather sensor data, recognize activities, and determine semantic labels without requiring external manual input. The electronic device automatically collects sensor data, the activity recognizer processor determines activities from this data, and the semantic labeler processor assigns labels based on activity-location mappings, making the system self-sufficient.
Solution Approach 2:
The patent replaces the mechanical manual labeling process with an automated electronic system. Instead of manual user input, the system uses sensor data collection, activity recognition algorithms, and automated mapping processes to determine semantic labels, substituting human effort with electronic processing and computational analysis.
2Productivity
If automated labeling is implemented, then productivity increases, but measurement precision of place semantics may deteriorate
Solution Approach 1:
The system uses observed mappings between activities and locations from sensor data as feedback to determine semantic labels. The activity recognizer processor continuously monitors sensor data, determines activities, and feeds this information back to the semantic place labeler processor, which uses the observed activity-location relationships to accurately assign semantic labels, ensuring precision through iterative feedback.
Solution Approach 2:
The system changes the parameters used for labeling from manual user input to objective sensor-based activity data. By measuring physical parameters (sensor data) and transforming them into activity classifications, the system achieves accurate semantic labeling through parameter transformation rather than subjective manual input.
3Measurement precision
If activity recognition is used to differentiate locations, then ability to distinguish similar places improves, but device complexity increases
Solution Approach 1:
The system uses a multi-functional processor that performs both activity recognition and semantic label determination. The activity recognizer processor and semantic place labeler processor work together as an integrated system, allowing the same electronic device to perform multiple functions (sensor data analysis, activity recognition, location tracking, and semantic labeling) without requiring separate specialized systems for each function.
Data Source
AI summary
A method and device for automatic semantic labeling of unlabeled places using activity recognition. A method includes determining at least one activity based on analyzing electronic device sensor data. Localization for the electronic device is performed to determine location for an unknown semantic place. An observed mapping between the at least one activity and the location for the unknown semantic place is determined. A typical mapping between the at least one activity and at least one semantic place is determined. Using the observed mapping and the typical mapping from one or more other electronic devices, the unknown semantic place is assigned with a semantic place label representing the at least one semantic place for identifying the unknown semantic place. A semantic place map is updated to include the semantic place label.