MicroCinema: A Divide-and-Conquer Approach for Text-to-Video Generation.
Yanhui WangJianmin BaoWenming WengRuoyu FengDacheng YinTao YangJingxu ZhangQi Dai Zhiyuan ZhaoChunyu WangKai QiuYuhui YuanXiaoyan SunChong LuoBaining GuoPublished in: CoRR (2023)
Keyphrases
- text generation
- natural language descriptions
- video sequences
- text detection
- video search
- video data
- multimedia documents
- video streams
- space time
- video frames
- news video
- multimedia
- text mining
- real time video
- real time
- video segments
- natural language generation
- video analysis
- free text
- text information
- database
- spatial and temporal
- video retrieval
- video images
- video database
- keywords
- closed captions
- multimedia search
- audio content
- video clips
- video content
- key frames
- temporal information
- generation process
- textual data
- textual information
- text data
- text retrieval
- tv programs
- online video
- multimedia data
- feature vectors
- natural language
- information retrieval