Let Them Choose What They Want: A Multi-Task CNN Architecture Leveraging Mid-Level Deep Representations for Face Attribute Classification.
Zhenduo ChenFeng LiuZhenglai ZhaoPublished in: ICIP (2021)
Keyphrases
- mid level
- multi task
- low level
- high level
- learning tasks
- feature selection
- low level features
- perceptual grouping
- visual cues
- event detection
- decision trees
- levels of abstraction
- action recognition
- machine learning
- visual concepts
- image classification
- face images
- supervised learning
- feature extraction
- data sets
- higher level
- visual information
- text classification
- superpixels
- multi class
- unsupervised learning
- information gain
- feature vectors
- training set