Collaborative Reasoning on Multi-Modal Semantic Graphs for Video-Grounded Dialogue Generation.
Xueliang ZhaoYuxuan WangChongyang TaoChenshuo WangDongyan ZhaoPublished in: EMNLP (Findings) (2022)
Keyphrases
- multi modal
- semantic concepts
- video search
- multi modality
- video sequences
- video shots
- visual concepts
- video content
- natural language
- audio visual
- image annotation
- video streams
- cross modal
- multiple modalities
- semantic search
- dialogue system
- fusing multiple
- video database
- video analysis
- video data
- multimedia
- video clips
- video frames
- image processing
- single modality
- feature selection