Context-LSTM: a robust classifier for video detection on UCF101.
Dengshan LiRujing WangPublished in: CoRR (2022)
Keyphrases
- video data
- activity detection
- event detection
- face detection and tracking
- multimedia
- video sequences
- video streams
- support vector machine classifier
- human actions
- partial occlusion
- video database
- multimedia data
- contextual information
- label noise
- training data
- real time
- video analysis
- learning algorithm
- video content
- object detection
- feature space
- human activities
- recurrent neural networks
- video frames
- class labels
- action recognition
- training samples
- discriminative classifiers
- dynamic background
- context aware
- person detection
- feature selection