MRT: Multi-modal Short- and Long-range Temporal Convolutional Network for Time-sync Comment Video Behavior Prediction.
Weihao ZhaoWeidong HeHao WangHaoyang BiHan WuChen ZhuTong XuEnhong ChenPublished in: LREC/COLING (2024)
Keyphrases
- multi modal
- long range
- short range
- semantic concepts
- video search
- space time
- conditional random fields
- video sequences
- video content
- video data
- convolutional network
- multimedia
- multiple modalities
- convolutional neural networks
- audio visual
- high dimensional
- video analysis
- video frames
- video retrieval
- video shots
- visual information
- mutual information
- hidden markov models
- humanoid robot
- image processing
- information retrieval