Towards End-to-End Generative Modeling of Long Videos with Memory-Efficient Bidirectional Transformers.
Jaehoon YooSemin KimDoyup LeeChiheon KimSeunghoon HongPublished in: CoRR (2023)
Keyphrases
- end to end
- memory efficient
- ad hoc networks
- multipath
- congestion control
- video sequences
- wireless ad hoc networks
- high bandwidth
- video data
- iterative deepening
- real time
- rate allocation
- admission control
- content delivery
- text localization and recognition
- integral image
- application layer
- video content
- video frames
- real world