A dynamic gesture recognition and prediction system using the convexity approach

详细信息    查看全文

• 作者：Pablo Barros ; ^a ; ^{barros@informatik.uni-hamburg.de} ; Nestor T. Maciel-Junior^b ; ^{ntmj@ecomp.poli.br} ; Bruno J.T. Fernandes^b ; ^{bjtf@ecomp.poli.br} ; Byron L.D. Bezerra^b ; ^{byronleite@ecomp.poli.br} ; Sergio M.M. Fernandes^b ; ^{smmf@ecomp.poli.br}
• 关键词：Gesture recognition ; Computer vision ; Features extraction ; Gesture prediction
• 刊名：Computer Vision and Image Understanding
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：155
• 期：Complete
• 页码：139-149
• 全文大小：2507 K
• 卷排序：155

文摘

Novel hand gesture feature extractor algorithm (Convexity Approach). Application of the Convexity Approach in different techniques, extending them (CLCS, CSURF). A novel hand gestures dataset. Use of the Convexity Approach with different temporal classifiers. Showing that the proposed approach works better than state-of-the-art models.