End-to-end Concept Word Detection for Video Captioning, Retrieval, and Question Answering
| Autor: | Gunhee Kim, Jongwook Choi, Youngjae Yu, Hyungjin Ko |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Closed captioning
FOS: Computer and information sciences Computer science business.industry Speech recognition Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition 020207 software engineering 02 engineering and technology Semantics computer.software_genre Visualization Knowledge extraction 0202 electrical engineering electronic engineering information engineering Question answering 020201 artificial intelligence & image processing Language model Artificial intelligence business computer Natural language processing Word (computer architecture) |
| Zdroj: | CVPR |
| DOI: | 10.48550/arxiv.1610.02947 |
| Popis: | We propose a high-level concept word detector that can be integrated with any video-to-language models. It takes a video as input and generates a list of concept words as useful semantic priors for language generation models. The proposed word detector has two important properties. First, it does not require any external knowledge sources for training. Second, the proposed word detector is trainable in an end-to-end manner jointly with any video-to-language models. To maximize the values of detected words, we also develop a semantic attention mechanism that selectively focuses on the detected concept words and fuse them with the word encoding and decoding in the language model. In order to demonstrate that the proposed approach indeed improves the performance of multiple video-to-language tasks, we participate in four tasks of LSMDC 2016. Our approach achieves the best accuracies in three of them, including fill-in-the-blank, multiple-choice test, and movie retrieval. We also attain comparable performance for the other task, movie description. Comment: In CVPR 2017. Winner of three (fill-in-the-blank, multiple-choice test, and movie retrieval) out of four tasks of the LSMDC 2016 Challenge. 22 pages |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|