| Presentation | 2023-03-02 [Poster Presentation] Gait Analysis Focusing on Operating Characteristics at Feature Points Detected by OpenPose Chinatsu Tanaka, Minoru Kuribayashi, Nobuo Funabiki, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A person's walking motion contains a variety of information, such as age and gender. Gait verification, which uses the individuality of a person's walking style to authenticate the individual, has been attracting attention. Conventional gait recognition methods include model-based methods that extract features by fitting a person model on an image, and appearency-based methods that extract features based on a time series of silhouette images. In the conventional walker identification, two types of methods have been studied: a model-based method that extracts features by fitting a person model on an image, and an appearance-based method that extracts features based on a time series of silhouette images. In particular, many model-based studies use skeletal estimation as a feature. However, since these studies use skeletal coordinate data estimated from video images as they are, the identification accuracy may deteriorate depending on the position of the person in the image. In this study, we propose a model that estimates the skeleton using OpenPose and identifies a person based on the difference between the target and predicted values using a model that predicts the skeletal movement of a typical person. The videos used are those shot with a fixed camera, and the distance traveled by the skeletal coordinate data in the image is used as the feature value. The effectiveness of the proposed method was demonstrated by comparing it with a method using actual skeletal coordinate data through simulation. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Gait Recognition / Human Pose Estimation / OpenPose / Machine Learning / LSTM / CNN |
| Paper # | EMM2022-83 |
| Date of Issue | 2023-02-23 (EMM) |
| Conference Information | |
| Committee | EMM |
|---|---|
| Conference Date | 2023/3/2(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Fukue culture hall |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Ryoichi Nishimura(NICT) |
| Vice Chair | Kotaro Sonoda(Nagasaki Univ.) / Masatsugu Ichino(Univ. of Electro-Comm.) |
| Secretary | Kotaro Sonoda(Kaishi Professional Univ.) / Masatsugu Ichino(Chiba Univ.) |
| Assistant | Tomoko Kajiyama(Hiroshima City Univ.) / Shieyuki Sakazawa(Osaka Inst. of Tech.) |
| 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) | [Poster Presentation] Gait Analysis Focusing on Operating Characteristics at Feature Points Detected by OpenPose |
| Sub Title (in English) | |
| Keyword(1) | Gait Recognition |
| Keyword(2) | Human Pose Estimation |
| Keyword(3) | OpenPose |
| Keyword(4) | Machine Learning |
| Keyword(5) | LSTM |
| Keyword(6) | CNN |
| 1st Author's Name | Chinatsu Tanaka |
| 1st Author's Affiliation | Okayama University(Okayama Univ.) |
| 2nd Author's Name | Minoru Kuribayashi |
| 2nd Author's Affiliation | Okayama University(Okayama Univ.) |
| 3rd Author's Name | Nobuo Funabiki |
| 3rd Author's Affiliation | Okayama University(Okayama Univ.) |
| Date | 2023-03-02 |
| Paper # | EMM2022-83 |
| Volume (vol) | vol.122 |
| Number (no) | EMM-412 |
| Page | pp.pp.84-88(EMM), |
| #Pages | 5 |
| Date of Issue | 2023-02-23 (EMM) |