Automatic textual annotation of video news based on semantic visual object extraction.
Nozha BoujemaaFrançois FleuretValérie GouetHichem SahbiPublished in: Storage and Retrieval Methods and Applications for Multimedia (2004)
Keyphrases
- object extraction
- visual concepts
- semantic labels
- textual descriptions
- news video
- multimedia
- cross media
- semantic concepts
- visual content
- visual and textual features
- web images
- video annotation
- visual features
- semantic annotation
- concept detectors
- keywords
- video content
- video search
- metadata
- image annotation
- visual information
- semantic information
- low level
- object segmentation
- video sequences
- video retrieval
- image segmentation
- image content
- video data
- multi modal
- high level
- key frames
- low level features
- active learning
- video frames
- news stories
- learning algorithm
- object recognition