Learning to Generate Text-Grounded Mask for Open-World Semantic Segmentation from Only Image-Text Pairs.
Junbum ChaJonghwan MunByungseok RohPublished in: CVPR (2023)
Keyphrases
- semantic segmentation
- weakly supervised
- superpixels
- single image
- text mining
- learning algorithm
- image content
- input image
- image features
- pixel level
- image segmentation
- information retrieval
- image representation
- conditional random fields
- image matching
- bounding box
- scene classification
- segmentation algorithm
- high resolution
- incomplete information
- image retrieval
- object class