MemBridge: Video-Language Pre-Training With Memory-Augmented Inter-Modality Bridge.
Jiahao YangXiangyang LiMao ZhengZihan WangYongqing ZhuXiaoqian GuoYuchen YuanZifeng ChaiShuqiang JiangPublished in: IEEE Trans. Image Process. (2023)
Keyphrases
- multi modal
- video sequences
- video data
- video streams
- language learning
- multimedia
- real time
- video analysis
- video retrieval
- data sets
- video database
- memory requirements
- memory space
- memory usage
- digital video
- video surveillance
- natural language
- training examples
- training samples
- spatio temporal
- test set
- video clips
- computer vision
- training set
- training phase
- visual cues
- moving objects
- training algorithm
- artificial neural networks
- supervised learning
- online learning
- temporal information
- video frames
- space time