MILES: Visual BERT Pre-training with Injected Language Semantics for Video-Text Retrieval.
Yuying GeYixiao GeXihui LiuJinpeng WangJianping WuYing ShanXiaohu QiePing LuoPublished in: ECCV (35) (2022)
Keyphrases
- text retrieval
- information retrieval
- document retrieval
- video sequences
- multimedia
- video content
- document collections
- image retrieval
- visual information
- video data
- retrieval systems
- inverted file
- cross language
- visual features
- space time
- query expansion
- low level
- multimedia data
- retrieval quality
- video retrieval
- video search
- automatic query expansion
- latent semantic indexing
- bag of visual words
- textual information retrieval
- medical image retrieval
- feature extraction
- visual concepts
- machine learning
- active learning
- training set
- natural language