Align2Ground: Weakly Supervised Phrase Grounding Guided by Image-Caption Alignment.
Samyak DattaKaran SikkaAnirban RoyKaruna AhujaDevi ParikhAjay DivakaranPublished in: ICCV (2019)
Keyphrases
- weakly supervised
- input image
- image data
- weakly labeled
- multiscale
- image retrieval
- superpixels
- bounding box
- single image
- image features
- keypoints
- region of interest
- image content
- image classification
- image regions
- image processing
- test images
- segmentation method
- object class
- low level
- image segmentation
- spatial information
- image representation
- object detectors