Two-Stage Polyphonic Sound Event Detection Based on Faster R-CNN-LSTM with Multi-Token Connectionist Temporal Classification.
In Young ParkHong Kook KimPublished in: INTERSPEECH (2020)
Keyphrases
- event detection
- composite events
- event recognition
- video event detection
- video analysis
- activity recognition
- video surveillance
- multimedia event detection
- musical instrument
- temporal segmentation
- text classification
- image classification
- space time
- neural network
- scan statistic
- machine learning
- feature selection
- feature vectors
- spatio temporal
- feature space
- video event
- computer vision
- spatial and temporal
- temporal databases
- sports video
- training data
- recurrent neural networks
- knn
- temporal information
- trecvid multimedia event detection
- active database management systems