ViT-ReT: Vision and Recurrent Transformer Neural Networks for Human Activity Recognition in Videos.
James WenselHayat UllahArslan MunirPublished in: CoRR (2022)
Keyphrases
- human activity recognition
- human activities
- neural network
- activity recognition
- recurrent neural networks
- fuzzy logic
- fault diagnosis
- feed forward
- visual surveillance
- vision system
- atomic actions
- human actions
- action recognition
- video sequences
- computer vision
- sensor data
- back propagation
- human computer interaction
- video data
- video frames
- video content
- video database
- object detection
- sensor networks
- event recognition
- data sets