Visual Content Recognition by Exploiting Semantic Feature Map with Attention and Multi-task Learning.
Rui-Wei ZhaoQi ZhangZuxuan WuJianguo LiYu-Gang JiangPublished in: ACM Trans. Multim. Comput. Commun. Appl. (2019)
Keyphrases
- visual content
- multi task learning
- multi task
- visual features
- video retrieval
- visual information
- image content
- image retrieval
- low level
- learning tasks
- image collections
- low level features
- feature extraction
- object recognition
- visual words
- semantic information
- high level
- key frames
- action recognition
- transfer learning
- image representation
- unsupervised learning
- data sets
- image classification
- multiscale
- information retrieval
- saliency map
- regression model
- eye movements
- e learning
- neural network