A dataset for Arabic text detection, tracking and recognition in news videos- AcTiV.
Oussama ZayeneJean HennebertSameh Masmoudi ToujRolf IngoldNajoua Essoukri Ben AmaraPublished in: ICDAR (2015)
Keyphrases
- text detection
- natural scene images
- news video
- video database
- object recognition
- video analysis
- human actions
- text information
- connected components
- urban scenes
- action recognition
- news stories
- video sequences
- multimedia databases
- low level features
- video data
- temporal information
- multimedia
- partial occlusion
- visual information
- image data
- feature extraction