InvPT++: Inverted Pyramid Multi-Task Transformer for Visual Scene Understanding.
Hanrong YeDan XuPublished in: CoRR (2023)
Keyphrases
- scene understanding
- multi task
- scene categorization
- object detection
- multi task learning
- object recognition
- learning tasks
- vision system
- d scene
- multi class
- video surveillance
- learning problems
- transfer learning
- feature selection
- image representation
- multiscale
- single image
- information gain
- image classification
- decision trees
- data mining