Time-Lag Aware Multi-Modal Variational Autoencoder Using Baseball Videos And Tweets For Prediction Of Important Scenes.
Kaito HirasawaKeisuke MaedaTakahiro OgawaMiki HaseyamaPublished in: ICIP (2021)
Keyphrases
- multi modal
- video search
- dynamic scenes
- audio visual
- sports video
- video data
- semantic concepts
- event recognition
- multi modality
- video scene
- cross modal
- image annotation
- social media
- video analysis
- uni modal
- video frames
- video sequences
- image segmentation
- optical flow
- single modality
- fusing multiple
- tv series
- computer vision