HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video Clips.
Antoine MiechDimitri ZhukovJean-Baptiste AlayracMakarand TapaswiIvan LaptevJosef SivicPublished in: ICCV (2019)
Keyphrases
- video clips
- m learning
- tv shows
- video segments
- mobile learning
- video database
- video collections
- video data
- video streams
- video content
- video frames
- multimedia documents
- video retrieval
- mobile phone
- higher education
- long video
- video material
- e learning
- key frames
- closed captions
- learning experience
- video dataset
- mobile applications
- distance education
- context aware
- learning technologies
- mobile devices
- video shots
- learning systems
- user experience
- keywords
- information retrieval
- computer vision
- temporal information
- topic segmentation
- video scene
- machine learning
- three dimensional
- video sequences
- video search
- digital video
- database
- search engine
- video analysis