End-to-End Subtitle Detection and Recognition for Videos in East Asian Languages via CNN Ensemble with Near-Human-Level Performance.
Yan XuSiyuan ShanZiming QiuZhipeng JiaZhengyang ShenYipei WangMengfei ShiEric I-Chao ChangPublished in: CoRR (2016)
Keyphrases
- end to end
- human level
- text localization and recognition
- artificial general intelligence
- machine intelligence
- convolutional neural network
- intelligent systems
- human level ai
- web intelligence
- congestion control
- admission control
- cognitive science
- video sequences
- video data
- real time
- human intelligence
- video frames
- cognitive psychology
- ai systems
- natural language
- computer vision