DeepActsNet: A deep ensemble framework combining features from face, hands, and body for action recognition.
Umar AsifDeval MehtaStefan von CavallarJianbin TangStefan HarrerPublished in: Pattern Recognit. (2023)
Keyphrases
- action recognition
- body parts
- video dataset
- computer vision
- bag of features
- image features
- spatial temporal
- view invariant
- bag of words
- human body
- human pose
- activity recognition
- action classification
- human movements
- deformable part models
- object detection
- low level
- human actions
- human detection
- recognition of human actions
- machine learning
- static images
- space time interest points
- mid level
- motion history images
- multiple features
- visual words
- video frames
- keypoints
- support vector machine
- spatio temporal
- training set
- feature space
- feature extraction
- face recognition