Coarse-to-Fine Semantic Alignment for Cross-Modal Moment Localization.
Yupeng HuLiqiang NieMeng LiuKun WangYinglong WangXian-Sheng HuaPublished in: IEEE Trans. Image Process. (2021)
Keyphrases
- coarse to fine
- cross modal
- multi modal
- multiresolution
- multiscale
- image registration
- hierarchical segmentation
- object detection
- visual similarity
- semantic concepts
- multimedia retrieval
- visual recognition
- dynamic programming
- multimedia databases
- image retrieval
- semantic information
- machine learning
- visual data
- low level features
- semantic similarity
- multimedia data
- multimedia information retrieval
- image classification
- low level
- feature space
- reinforcement learning
- high level
- metadata