| Presentation | 2017-11-07 [Invited Talk] Researches on Impressiveness Multimedia Contents Toshihiko Yamasaki, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Due to the advancement in machine learning and pattern recognition technologies such as deep neural networks, recognition of “what” has been made very accurate. For instance, object detection and recognition in natural images is a task of understanding “what’s in this picture?” Artificial intelligence developed for Go and Shogi games is also designed for “what’s next?” We believe that understanding “how” and “why” in addition to “what” will be of great importance in the future. For this purpose, understanding humans’ internal state such as sentiment/emotion and intent would also be important. From this point of view, our research group has been working on impression prediction and analysis of multimedia content. It is believed that impressive contents can be created only by those who are talented and experienced so far. However, our research group has revealed that with enough amount of data, proper labels, and proper machine learning techniques, predicting how impressive the contents are is possible. It is also possible to give feedbacks to creators on how to make the contents more impressive. In my presentation, I would like to introduce some of our representative works such as oral presentation analysis, recognition rate and favorableness score prediction of TV commercials, and so on. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Big Multimedia Data / Impression Analysis / Machine Learning / Pattern Recognition / Deep Neural Networks / Attractiveness |
| Paper # | EMM2017-63 |
| Date of Issue | 2017-10-30 (EMM) |
| Conference Information | |
| Committee | EMM |
|---|---|
| Conference Date | 2017/11/6(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Kagoshima Univ. (Inamori Academy) |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Multimedia Fusion, Content Processing, Multimedia Retrieval, Digital Watermarking, and Related Topics |
| Chair | Keiichi Iwamura(TUS) |
| Vice Chair | Hirohisa Hioki(Kyoto Univ.) / Minoru Kuribayashi(Okayama Univ.) |
| Secretary | Hirohisa Hioki(Shizuoka Univ.) / Minoru Kuribayashi(Tokyo Metro. Univ.) |
| Assistant | Kan Hyonho(NIT, Tokyo) / Harumi Murata(Chukyo Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Enriched MultiMedia |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | [Invited Talk] Researches on Impressiveness Multimedia Contents |
| Sub Title (in English) | Attractiveness Computing using Big Multimedia Data |
| Keyword(1) | Big Multimedia Data |
| Keyword(2) | Impression Analysis |
| Keyword(3) | Machine Learning |
| Keyword(4) | Pattern Recognition |
| Keyword(5) | Deep Neural Networks |
| Keyword(6) | Attractiveness |
| 1st Author's Name | Toshihiko Yamasaki |
| 1st Author's Affiliation | The University of Tokyo(UTokyo) |
| Date | 2017-11-07 |
| Paper # | EMM2017-63 |
| Volume (vol) | vol.117 |
| Number (no) | EMM-282 |
| Page | pp.pp.45-45(EMM), |
| #Pages | 1 |
| Date of Issue | 2017-10-30 (EMM) |