activity recognition from user-annotated acceleration data ling
TRANSCRIPT
Activity Recognition from
User-Annotated
Acceleration Data
Ling Bao and Stephen S. Intille
Presented by: Hong Lu
Key Questions
• Can low cost wearable sensors be used for robust, real- time recognition of activity?
• Can training data be acquired from the end user without researcher supervision?
• Does recognition require user-specific training data?
• Do more sensors improve recognition?
Data Collection
• 13 ♂ + 7♀ = 20 subjects , age from 17 to 48
• 20 everyday activities
• Subjects unsupervised when generating own training data, both in and outside the lab
• What’s the problem of typical laboratory data? WHY?
Often data in lab is collected from researchers as subjects
Lab environments may restrict activity, simplifying recognition !
Making researchers to label training examples does not scale
Recognition rates highly depended on how data is collected95.6% (laboratory data)
VS66.7% (naturalistic settings)
Data Collection
• What’s an accelerometer ?
• An accelerometer is a device that measures the vibration, or acceleration of motion of a structure.
Why Accelerometer ?
• Many daily activities involve repetitive physical motion of the body or specific postures• E.g. Walking, Running, Scrubbing, Vacuuming
• Low cost, tiny, energy efficient • Watch
• Phone, mp3 player
• Camera
• computer
• Game controller, the wii remote
Sensor Placement
• 5 wireless sensors
Right hip
Wrist
upper arm
Ankle
Thigh
• Shack to synchronize
Raw Data
Features
• Why we need them ?
• Summarize the data bin
• Capture useful information
• What is the desired characteristics of a good feature ?
• removing irrelevant noise
• keeping relevant attributes to tell the difference
• easy to compute
?
Features
• 512 sample windows (6.7s ?), 50% window overlap
• Features: • Mean
• Energy
• Frequency-domain entropy
• Correlation Between x, y accelerometer axes each board Between all pair wise combinations of axes on different boards
Classifiers
• Tested on decision table, nearest neighbor ( IBL), C4.5 decision tree, and naïve Bayesian classifiers
• Machine Learning Toolkit (Witten & Frank, 1999)
Training
• Method 1: User-specific training
• Train on activity sequence data for each subject
• Test on obstacle course data for that subject
• Method 2: Leave-one-subject out training
• Train on activity sequence and activity data for all subjects but one
• Test on obstacle course data for left out subject
• Average for all 20 subjects
Results
• C45 Decision tree wins
• It shows
• User-specific training: 71.6 ±7.4
• Leave-one-subject-out training: 84.3 ±5.2
• Why?
• Commonalities between people may be more significant than individual variations
• Larger training set
Result
• Overall, promising • Data collected by subjects themselves without
supervision • Data collected both in and outside of laboratory
setting
• Poorer performance results when… • Activities involve less physically characteristic
movements , Activities involve little motion or standing still
• Activities involve similar posture/movement (e.g. watching TV, sitting and relaxing)
The dark side
• The more sensors you placed, the higher accuracy you may achieved, but …
• cost
• you look weird
• hard to deploy
• more computational horse power
Accelerometer Discriminatory Power
• Tested C4.5 classifier with using subsets of accelerometers:
• Hip, wrist, arm, ankle, thigh, thigh and wrist, hip and wrist
• Best single performers:
• Thigh (-29.5%)
• Hip (-34.1%)
• Ankle(-37%)
Accelerometer Discriminatory Power
• With only two accelerometers get good performance:
• Thigh and wrist (-3.3% compared with all 5)
• Hip and wrist (-4.8% compared with all 5)
Overview
• The study • Activity recognition: 20 household activities
• Sensors: 5 non-wired accelerometers
• Data: participants labeled own data
• Result • Good performance with decision tree classifier
• Subject-specific training data for some activities may not be required
• Reasonable accuracy can be achieved with only 2 of 5 accelerometers
Thank you!
The End
For some slides, I used content of Emmanuel MunguiaTapia’s presentation