Massive-scale learning of image and video semantic concepts.
John R. SmithLiangliang CaoNoel C. F. CodellaMatthew L. HillMichele MerlerQuoc-Bao NguyenE. PringRosario A. Uceda-SosaPublished in: IBM J. Res. Dev. (2015)
Keyphrases
- semantic concepts
- low level
- image collections
- image data
- visual concepts
- image content
- image features
- key frames
- image retrieval
- massive scale
- semantic gap
- semantic concept detection
- image representation
- image classification
- video sequences
- active learning
- multi modal
- feature extraction
- action recognition
- image segmentation
- image regions
- multimedia
- multimedia data
- video content
- concept detection
- video data
- visual content
- image annotation
- image matching
- keypoints
- real time
- image database
- feature vectors
- multiscale
- metadata