Loading...
Please wait, while we are loading the content...
Similar Documents
Image-Based Emotion Retrieval Approach with Multi-machine Learning Schemes
| Content Provider | Semantic Scholar |
|---|---|
| Author | Chang, Jae-Khun Ryoo, Seung-Taek |
| Copyright Year | 2012 |
| Abstract | With the development of cognitive abilities of people, the human’s individual emotion would be very different styles in accordance with his environment and condition. Therefore, the human’s emotions to see and feel the same image can be different and this results in different images to see. In this paper, we construct a new style emotion model and propose a new emotion retrieval method from the image using the model. The emotion retrieval method in this research is composed of two phases. First, feature points to represent the image are extracted. Feature points are hue, saturation, frequency information, and circularity of image which represents the shape of the object. The extracted feature values are used as inputs of machine learning scheme in second phase. In machine learning scheme, learning and testing are performed with the typical three learning schemes. Through these machine learning schemes, the human’s emotion is extracted when to see a new image, and a new paradigm is provided to apply many different fields. |
| File Format | PDF HTM / HTML |
| DOI | 10.1007/978-94-007-5064-7_26 |
| Alternate Webpage(s) | https://page-one.springer.com/pdf/preview/10.1007/978-94-007-5064-7_26 |
| Alternate Webpage(s) | https://doi.org/10.1007/978-94-007-5064-7_26 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
