Towards subject independent continuous sign language recognition: A segment and merge approach

作者：

Highlights：

• Variations in sign language are examined to develop a signer independent system.

• A 4-channel phoneme-based approach is used.

• Continuous sentence is segmented into sign or movement epenthesis sub-segments.

• Sign sub-segments are merged and recognized with a two-layer CRF.

• Novel decoding scheme is proposed for the semi-Markov CRF used in the 2-layer CRF.

摘要

Highlights•Variations in sign language are examined to develop a signer independent system.•A 4-channel phoneme-based approach is used.•Continuous sentence is segmented into sign or movement epenthesis sub-segments.•Sign sub-segments are merged and recognized with a two-layer CRF.•Novel decoding scheme is proposed for the semi-Markov CRF used in the 2-layer CRF.

论文关键词：Gesture recognition,Sign language recognition,Signer independence,Bayesian network,Conditional random field (CRF),Support vector machine (SVM),Semi-Markov CRF,Hidden Markov model (HMM)

论文评审过程：Received 2 June 2012, Revised 12 September 2013, Accepted 21 September 2013, Available online 30 September 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.09.014