Human Activity Recognition From Multi-modal Wearable Sensor Data Using Deep Multi-stage LSTM Architecture Based on Temporal Feature Aggregation.
Tanvir MahmudShaimur Salehin AkashShaikh Anowarul FattahWei-Ping ZhuM. Omair AhmadPublished in: MWSCAS (2020)
Keyphrases
- multi modal
- multistage
- sensor data
- human activity recognition
- human activities
- activity recognition
- sensor networks
- wearable sensors
- ubiquitous computing
- lifelog
- heterogeneous sensor networks
- single stage
- smart environments
- visual surveillance
- sensor measurements
- dynamic programming
- data streams
- human computer interaction
- wireless sensor networks
- temporal information
- raw sensor data
- optimal policy
- high dimensional
- feature vectors
- daily life
- low cost
- action recognition
- video data
- data sets
- content based retrieval
- user interface
- ambient intelligence