BOSS: Bottom-up Cross-modal Semantic Composition with Hybrid Counterfactual Training for Robust Content-based Image Retrieval.
Wenqiao ZhangJiannan GuoMengze LiHaochen ShiShengyu ZhangJuncheng LiSiliang TangYueting ZhuangPublished in: CoRR (2022)
Keyphrases
- cross modal
- image retrieval
- multi modal
- semantic information
- semantic content
- high level
- multimedia
- active learning
- feature space
- multimedia retrieval
- visual recognition
- multimedia databases
- multimedia data
- metadata
- multimedia content
- low level features
- action recognition
- search engine
- relevance feedback
- high dimensional
- feature extraction