Gesture Recognition with a 3-D Accelerometer. Wu, J H, Pan, G, Zhang, D Q, Qi, G D, & Li, S J 5585:25–38.
Gesture Recognition with a 3-D Accelerometer [link]Paper  doi  abstract   bibtex   
Gesture-based interaction, as a natural way for human-computer interaction, has a wide range of applications in ubiquitous computing environment. This paper presents an acceleration-based gesture recognition approach, called FDSVM (Frame-based Descriptor and multi-class SVM), which needs only a wearable 3-dimensional accelerometer. With FDSVM, firstly, the acceleration data of a gesture is collected and represented by a frame-based descriptor, to extract the discriminative information. Then a SVM-based multi-class gesture classifier is built for recognition in the nonlinear gesture feature space. Extensive experimental results on a data set with 3360 gesture samples of 12 gestures over weeks demonstrate that the proposed FDSVM approach significantly outperforms other four methods: DTW, Naive Bayes, C4.5 and HMM. In the user-dependent case, FDSVM achieves the recognition rate of 99.38% for the 4 direction gestures and 95.21% for all the 12 gestures. In the user-independent case, it obtains the recognition rate of 98.93% for 4 gestures and 89.29% for 12 gestures. Compared to other accelerometer-based gesture recognition approaches reported in literature FDSVM gives the best resulrs for both user-dependent and user-independent cases.
@article{wuGestureRecognition3D2009,
  title = {Gesture {{Recognition}} with a 3-{{D Accelerometer}}},
  volume = {5585},
  issn = {0302-9743},
  url = {https://www.thomsoninnovation.com/tip-innovation/%5Cnhttps://www.thomsoninnovation.com/tip-innovation/recordView.do?datasource=WOK&category=LIT&selRecord=1&totalRecords=1&databaseIds=WOS&idType=uid/recordid&recordKeys=000270444600003/WOS:000270444600003},
  doi = {10.1007/978-3-642-02830-4_4},
  abstract = {Gesture-based interaction, as a natural way for human-computer interaction, has a wide range of applications in ubiquitous computing environment. This paper presents an acceleration-based gesture recognition approach, called FDSVM (Frame-based Descriptor and multi-class SVM), which needs only a wearable 3-dimensional accelerometer. With FDSVM, firstly, the acceleration data of a gesture is collected and represented by a frame-based descriptor, to extract the discriminative information. Then a SVM-based multi-class gesture classifier is built for recognition in the nonlinear gesture feature space. Extensive experimental results on a data set with 3360 gesture samples of 12 gestures over weeks demonstrate that the proposed FDSVM approach significantly outperforms other four methods: DTW, Naive Bayes, C4.5 and HMM. In the user-dependent case, FDSVM achieves the recognition rate of 99.38\% for the 4 direction gestures and 95.21\% for all the 12 gestures. In the user-independent case, it obtains the recognition rate of 98.93\% for 4 gestures and 89.29\% for 12 gestures. Compared to other accelerometer-based gesture recognition approaches reported in literature FDSVM gives the best resulrs for both user-dependent and user-independent cases.},
  journaltitle = {Ubiquitous Intelligence and Computing, Proceedings;5585: 25-38 2009},
  date = {2009},
  pages = {25--38},
  author = {Wu, J H and Pan, G and Zhang, D Q and Qi, G D and Li, S J},
  file = {/home/dimitri/Nextcloud/Zotero/storage/GQAZSX9Z/Wu et al. - 2009 - Gesture recognition with a 3-d accelerometer.pdf}
}
Downloads: 0
About
Documentation
Keywords
Search
Users
Terms and Conditions of Use
Privacy Policy
Sitemap
© BibBase.org 2021