VQS: Linking Segmentations to Questions and Answers for Supervised Attention in VQA and Question-Focused Semantic Segmentation.
Chuang GanYandong LiHaoxiang LiChen SunBoqing GongPublished in: CoRR (2017)
Keyphrases
- semantic segmentation
- answer questions
- label transfer
- supervised methods
- correct answers
- weakly supervised
- question answer pairs
- street scenes
- community question answering
- superpixels
- conditional random fields
- object segmentation
- image database
- image segmentation
- object categories
- question answering
- named entities
- ground truth
- object class
- scene classification
- semi supervised
- visual attention
- medical images
- pascal voc
- object classes
- graphical models
- object detection
- topic models
- supervised learning
- bag of words
- video sequences
- machine learning