Loading...
Please wait, while we are loading the content...
Similar Documents
Application of Deep Learning in Video Question Answering System
| Content Provider | Scilit |
|---|---|
| Author | Pandya, Mansi Parekhji, Arnav Shahane, Aniket Chavan, Palak V. Mangrulkar, Ramchandra S. |
| Copyright Year | 2021 |
| Description | Recently, a research area of machine learning that has grown in popularity is Video Question Answering (VQA). A lot of models based on this concept are used for static images but very few models incorporate videos. The VQA's fundamental purpose is to develop a state-of-the-art multi-task model that uses Video Question Answering and Affective routes, and both of these routes work simultaneously to produce an intuitive answer. The main tasks carried out are perceiving emotions from frames of a video using a pretrained Convolutional Neural Network (CNN), and these perceived emotion labels are sent to token-based, frame-based and integrated attention systems and also used to achieve an image caption which is later used to produce a relevant answer. Significant zones of the video are targeted using visual features, the question asked and the emotion labels by the attention model. Thus, machines can be made to understand the continuous and changing visual scenes of the world with the help of this model. Book Name: Design of Intelligent Applications Using Machine Learning and Deep Learning Techniques |
| Related Links | https://api.taylorfrancis.com/content/chapters/edit/download?identifierName=doi&identifierValue=10.1201/9781003133681-23&type=chapterpdf |
| Ending Page | 372 |
| Page Count | 20 |
| Starting Page | 353 |
| DOI | 10.1201/9781003133681-23 |
| Language | English |
| Publisher | Informa UK Limited |
| Publisher Date | 2021-07-07 |
| Access Restriction | Open |
| Subject Keyword | Book Name: Design of Intelligent Applications Using Machine Learning and Deep Learning Techniques Artificial Intelligence Model Video Question Question Answering Frames Vqa Neural Machine Emotion |
| Content Type | Text |
| Resource Type | Chapter |
