| Presentation | 2002/7/11 A Study on the Extraction of Facial Parameters Based on the Specification of MPEG-4 Yosuke YAMADA, Naofumi AOKI, Tsuyoshi YAMAMOTO, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We have studied on the synthesis of the facial animation based on the specification of MPEG-4, the international standard of video codec. However, in synthesizing facial expression animation from actual facial images, it is hard to extract the structual features, such as the position and displacement of the feature points. This study focuses the expression recognition of the facial images by using higher-order local autocorrelation features and multiresolution images, instead of extracting such structural features. And we investigate the possibility of a method which performs the synthesis of animation by using prepared facial parameters, according to the recognition results. Higher-order local autocorrelation features is the higher-order extension of the autocorrelation function which is shift-invariant, and the range of the displacements is restricted within a 3 × 3 region, the center of which is the reference point. Instead of finding the best spacial resolution for the recognition of facial images, we generate multiresolution images by using wavelet transform, and extract higher-order local antocorrelation features from all multiresolution images. In the experiment, we photograph some images of subject's facial expression (neutral, joy, and angry) and create the histogram of higher-order local autocorrelation features extracted from each expression. And we investigate the effectivity of higher-order local autocorrelation features for the facial expression recognition. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Facial Expression recognition / Higher-order local autocorrelation features / Multiresolution images / MPEG-4 |
| Paper # | MVE2002-24 |
| Date of Issue |
| Conference Information | |
| Committee | MVE |
|---|---|
| Conference Date | 2002/7/11(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Media Experience and Virtual Environment (MVE) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Study on the Extraction of Facial Parameters Based on the Specification of MPEG-4 |
| Sub Title (in English) | |
| Keyword(1) | Facial Expression recognition |
| Keyword(2) | Higher-order local autocorrelation features |
| Keyword(3) | Multiresolution images |
| Keyword(4) | MPEG-4 |
| 1st Author's Name | Yosuke YAMADA |
| 1st Author's Affiliation | Graduate School of Engineering, Hokkaido University() |
| 2nd Author's Name | Naofumi AOKI |
| 2nd Author's Affiliation | Graduate School of Engineering, Hokkaido University |
| 3rd Author's Name | Tsuyoshi YAMAMOTO |
| 3rd Author's Affiliation | Graduate School of Engineering, Hokkaido University |
| Date | 2002/7/11 |
| Paper # | MVE2002-24 |
| Volume (vol) | vol.102 |
| Number (no) | 219 |
| Page | pp.pp.- |
| #Pages | 4 |
| Date of Issue | |