Video text recognition using category-dependent feature extraction based on feature compensation.
Minoru MoriMinako SawakiNorihiro HagitaPublished in: Systems and Computers in Japan (2005)
Keyphrases
- feature extraction
- feature vectors
- video streams
- text retrieval
- video sequences
- feature representation
- video content
- feature fusion
- real time
- feature set
- text detection
- video search
- key frames
- natural language descriptions
- feature space
- preprocessing
- image features
- video database
- feature extractor
- multimedia documents
- video analysis
- image feature extraction
- video frames
- image processing
- video data
- wavelet transform
- text documents
- event detection
- space time
- multimedia
- video retrieval
- news video
- preprocessing step
- text mining
- video segments
- feature selection
- face recognition
- keywords
- object recognition
- feature values
- natural language processing
- text categorization
- video clips
- object categories
- web documents
- linear discriminant analysis
- semantic information