Thai sign language translation system using upright speed-up robust feature and c-means clustering
Sign language is an important communication tool for the deaf. In this paper, we build a dynamic hand gesture translation system with video caption without prior hand region segmentation. In particular, we utilize the upright speed-up robust feature (U-SURF) and the fuzzy C-means (FCM) to find a mat...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference Proceeding |
| Published: |
2018
|
| Subjects: | |
| Online Access: | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84867588322&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/51507 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Chiang Mai University |
| Summary: | Sign language is an important communication tool for the deaf. In this paper, we build a dynamic hand gesture translation system with video caption without prior hand region segmentation. In particular, we utilize the upright speed-up robust feature (U-SURF) and the fuzzy C-means (FCM) to find a matched word. We compared the result with that from string grammar hard C-means (sgHCM). This Thai sign language translation system was tested on 42 words. The total number of video sequences used in the experiment is 1470. The best correct classification rate on the signer-dependent blind test set using the FCM is approximately 47 to 73%, whereas that of signer-semi-independent set is around 30 to 40%. The best blind test result for the signer-independent experiment is around 24 to 30%. However, the correct classification rate from the sgHCM is higher than that from the FCM. The best result for the signer-dependent experiment is around 97 to 99%, whereas that of the signer-semi-independent is around 64 to 65%. The correct classification rate of the signer-independent experiment is around 53 to 54%. © 2012 IEEE. |
|---|