A comparative study on wearables and single-camera video for upper-limb out-of-thelab activity recognition with different deep learning architectures.
Mario Martínez-ZarzuelaDavid González OrtegaMíriam Antón-RodríguezFrancisco Javier Díaz PernasHenning MüllerCristina Simon-MartinezPublished in: CoRR (2024)
Keyphrases
- activity recognition
- single camera
- deep learning
- human activities
- multiple cameras
- event detection
- video surveillance
- action recognition
- unsupervised learning
- visual surveillance
- field of view
- machine learning
- video sequences
- video data
- multi camera
- dynamic scenes
- partial occlusion
- activity analysis
- object tracking
- multi view
- image classification
- active learning
- moving objects
- object recognition
- image sequences