SeCG: Semantic-Enhanced 3D Visual Grounding via Cross-modal Graph Attention.
Feng XiaoHongbin XuQiuxia WuWenxiong KangPublished in: CoRR (2024)
Keyphrases
- cross modal
- semantic space
- multi modal
- visual similarity
- semantic concepts
- multimedia retrieval
- perceptual information
- image retrieval
- semantic information
- visual data
- semantic similarity
- multimedia databases
- visual recognition
- information retrieval
- visual attention
- higher level
- visual concepts
- video sequences
- multimedia data
- semantic content
- image classification
- co occurrence
- high level
- multimedia