Utilization of texture, contrast and color homogeneity for detecting and recognizing text from video frames.
Serhat TekinalpA. Aydin AlatanPublished in: ICIP (3) (2003)
Keyphrases
- video frames
- text regions
- natural scene images
- scene text
- text detection
- video sequences
- text extraction
- video segmentation
- video data
- color features
- color images
- color distribution
- color information
- texture information
- text information
- news video
- foreground objects
- texture features
- video clips
- saliency map
- image regions
- foreground and background regions
- complex background
- natural scenes
- foreground regions
- video content
- temporal coherence
- bounding box
- video dataset
- key frames
- input image
- connected components
- successive frames
- video shots
- video retrieval
- gray level
- video images
- monocular video sequences
- image sequences
- metadata