SemAug: Semantically Meaningful Image Augmentations for Object Detection Through Language Grounding.
Morgan HeislerAmin Banitalebi-DehkordiYong ZhangPublished in: CoRR (2022)
Keyphrases
- semantically meaningful
- object detection
- object level
- image data
- image features
- single image
- multiscale
- image retrieval
- image content
- template matching
- input image
- pixel level
- image classification
- keypoints
- higher order
- visual patterns
- image representation
- computer vision
- segmentation method
- image annotation
- target object
- bounding box
- image segmentation