Gesture recognition based on Global Template DTW for Chinese Sign Language

Abstract

This paper proposes global template dynamic time warping (GTDTW) algorithm for gesture recognition with the wearable gloves. The method is applied to both isolated and continuous gesture recognition. A gesture segmentation system based on GTDTW is also proposed for continuous gesture recognition. The global template is obtained based on statistical methods. For a defined gesture, states which have a large proportion are selected as important states. They form the global template for the defined gesture. Global template more fully expresses the characteristics of the defined gesture which can improve gesture recognition rate. Global template also has a smaller length than normal template of Dynamic Time Warping (DTW) so that time consumption of GTDTW is low and gesture recognition system has a better real-time performance. Experimental evaluations on both isolated and continuous gesture recognition show the effectiveness of the proposed method. The time consumption is obviously reduced and recognition rate is improved that up to 98.8% for isolated gesture recognition. For continuous gesture recognition, the proposed method has high segmentation rate and recognition rate is up to 95.6%.

Keywords

Global template dynamic time warping wearable gloves isolated gesture recognition continuous gesture recognition

1 Introduction

Gesture recognition is a hot topic in human-computer interaction. It is mainly based on image recognition or based on wearable devices [1]. Gesture recognition is divided into isolated gesture recognition and continuous gesture recognition. The methods of isolated gesture recognition usually are Dynamic Time Warping (DTW), Hidden Markov Model (HMM), neural network method and so on [2, 3]. DTW is widely used in gesture recognition to deal with the temporal data sequence. It can solve problem of matching similarity between temporal data sequence with different length. The basic process of DTW in gesture recognition is to select templates of defined gestures. Then the similarity is calculated between gesture sequence and different templates to choose highest similarity as recognition result. DTW is easy to implement but the performance is usually not ideal because of high time consumption and no-good template [4, 5]. The general way to select a DTW template is to select a sample as the template. But it is not good because there are spatial-temporal variability and noises between gestures sequence made by different persons. The general templates cannot fully express the characteristics of gesture so that recognition rate is low. The length of general templates usually equal to the length of the gesture. It is large so that time complexity of DTW is high. Some researchers improve the way to find optimal path [6] or are committed to improving the way of selecting templates [7]. They suggest that select multiple samples to find average as templates or select multiple samples as multiple templates for improving gesture recognition performance [8]. The recognition rate is improved but time consumption increases because more templates are used [9]. HMM is popular used in gesture recognition now because it is suitable for processing temporal data sequence, easy to add and modify gesture libraries [10, 11]. But it needs many samples to train model and time consumption is higher than DTW [12]. Recently, with the development of deep learning, neural network methods used for gesture recognition are growing up [13 –15]. For processing temporal data sequence, the recurrent neural network (RNN) is proposed and has a great effect [16]. However, the complexity of the neural network method cannot be ignored [17]. The neural network model requires more data to ensure the recognition performance. And in general, increasing the number of data can increase the complexity of training the NN model. The number of hidden layers is also not easy to determine. There will be the problem of gradient explosion, gradient disappearance and over-fitting if the number of hidden layer is too much. Recognition performance is not ideal. Some researchers suggest that DTW combined with HMM or neural network combined with HMM to improve together [18 –21]. But the weakness still cannot be completely eliminated.

For continuous gesture recognition, process of gesture segmentation is main concern. The task of gesture segmentation is to extract meaningful gesture from continuous gesture sequence [22 –25]. In other words, it needs to find the start points and the end points of gestures pattern. This is considered a difficult process because gestures have segmentation ambiguities and spatial-temporal variability. Most existing methods need to first implement gesture segmentation and then implement the recognition [26]. Thus, there is an unavoidable time delay between them which is not appropriate for continuous gesture recognition. To solve this problem, some researchers have used the HMM because it can model the spatial and temporal characteristics of gestures [27]. Kim et al. use forward spotting accumulative HMM for gesture segmentation but the method needs a model of no-gesture. The no-gesture is not defined gesture. There are many kinds of no-gestures so that it is complicated to train the models of them [28]. Deng and Tsui use multi-size windows HMM to alleviate the problem of training the model of no-gesture [29]. There are also researchers who are committed to discussing gesture segmentation based on DTW [30]. Li and Greenspan analyze all candidates paths obtained based on DTW [31] and choose the best gesture candidate. Some other methods are proposed for gesture segmentation to find the sudden change points as the start points and the end points [32, 33]. For example, find energy or speed sudden change points as basis of gesture segmentation. Most of them need to assume that the start points are similar to the end points or need appropriate thresholds to define sudden change points [34 –36]. But when thresholds are inappropriate or assumptions are not applicable, the recognition and segmentation performance may be influenced.

In this paper, GTDTW is proposed to improve the gesture recognition performance based on DTW. It is applied to both isolated and continuous gesture recognition. The global template can be more comprehensive expression of the characteristics of gestures which improves gesture recognition rate. The GTDTW reduce the size of the input sequence to feed into DTW so that time consumption of the proposed method is reduced. A gesture segmentation system based GTDTW is also proposed for continuous gesture recognition. The system implements gesture segmentation and recognition simultaneous and it has a well real-time performance. Combined with a novel gesture length threshold, the system can alleviate the influence of no-gestures.

The rest of this paper is organized as follows. The next section reviews details of the proposed method including GTDTW and gesture segmentation system. Section 3 describes experiments and results. Section 4 outlines conclusion.

2 Gesture recognition and GTDTW

In this paper, the GTDTW algorithm is proposed to use in gesture recognition, including isolated gesture recognition and continuous gesture recognition. The global templates show the most likely states of each defined gesture. The global templates fully express the characteristics of defined gestures which can improve gesture recognition performance. The time complexity of DTW is O (N²) where N is length of template. Global templates are much smaller than the normal DTW templates of other methods. The time consumption is reduced based on GTDWT. The real-time performance of gesture recognition is ensured based on the proposed method. In addition, the paper proposes a novel gesture segmentation system based on GTDTW combined with the length characteristics of different defined gestures. The basic idea of the proposed system is to observe the change of similarity between current gesture sequence and different templates and find the start point and the end point of gestures. The proposed method is real-time gesture segmentation which can directly output the recognition results after segmentation.

2.1 The global template

The global template can be obtained from samples for each defined gesture. Firstly, the gesture data needs to be discretized and K-means clustering algorithm is used for discretization. The clusters number defined as M and the gesture data at each moment is represented by an integer after discretization. The integer denotes the cluster that the gesture data belongs to. An illustration of the discretized gesture sequence is showed in Fig. 1. The number 1 in Fig. 1 indicates that gesture data obtained by sensors at these moments belongs to the same cluster which is the first cluster. The number 2 indicates that gesture data obtained by sensors at these moments belongs to the same cluster which is the second cluster. For all samples, all possible states of discretized gesture data are from 1 to M for integers which defined as observation state set S. S is showed as (1) $S = {s_{i} = i, i = 1, 2, . . ., M},$ (1) where s_i denotes i^th possible state. For a gesture, the change range of hand is limited. For example, the limited change range of index finger is showed in Fig. 2. Regardless of little number of noise points produced due to device and environmental problems, types of states that appear in a gesture sequence is limited as shown in Fig. 3.

Fig.1

An illustration of the discretized gesture sequence.

Fig.2

An illustration of limited change range of index finger.

Fig.3

An illustration of the types of states that (a) is 4, (b) is 3 and (c) is 2.

For a defined gesture g, different states s_i could be defined as key frame and sub-key frame according to the proportion θ_i of s_i in all samples. Considering that the total number of states is M, average occurrence probability for each state is 1/M. If the proportion of s_i is bigger than 1/M, s_i can better express the characteristics of gesture g than other states and define s_i as key frame of gesture g. Through experimental statistics, 0.1/M is defined as the threshold of sub-key frame. In other words, define key frame threshold as θ_key and sub-key frame threshold as θ_{sub_key} showed as (2) and (3) $θ_{key} = 1 / M,$ (2) $θ_{sub_key} = 0.1 / M .$ (3) If θ_i meets Eq.(4), s_i will be defined as key frame. If θ_i meets Eq.(5), s_i will be defined as sub-key frame. If θ_i meets Eq.(6), s_i will be defined as noise points. $θ_{key} < θ_{i} < 1,$ (4) $θ_{sub_key} < θ_{i} < θ_{key},$ (5) $0 < θ_{i} < θ_{sub_key} .$ (6)

For a defined gesture g, the global template is made up of key frames and sub-key frames which is defined as Y_g. The length of global template is just a single digit so that the length of the input to GTDTW is shorter and the time consumption of DTW is reduced. The function of sub-key frames is to ensure distinguishing gestures when part of key frames is repeated for different defined gestures.

2.2 Isolated gesture recognition

Isolated gesture recognition is to recognize individual gesture data, but continuous gesture recognition is to segment continuous gesture sequence and then recognize the segmented gesture data.

The similarity between gesture sequence and global templates Y_g is defined as D_g and calculated based on GTDTW. Define the weight of key frame as W_key and the weight of sub-key frame as W_{sub_key}. They are showed as Eqs.(7) and (8) $W_{key} = θ_{key},$ (7) $W_{sub_key} = θ_{sub_key} .$ (8) The similarity D_g is represented as Eq.(9) $D_{g} = {X, Y_{g}},$ (9) where the initial value of D_g is 0 and D_g increases W_key when key frames appear in gesture sequence X. D_g increases W_{sub_key} when sub-key frames appear and does not increase when noise points appear. The recognition result of gesture sequence X is defined as T and showed as Eq.(10) $T = \underset{g}{argmax} (D_{g}) .$ (10)

2.3 Continuous gesture recognition

For Continuous gesture recognition, a real-time gesture segmentation method is essential. The paper proposes a gesture segmentation system based GTDTW combined with the length characteristics of different defined gestures.

2.3.1 Length characteristics of gesture

For defined gesture g, its length characteristics is defined as length threshold L_g. It represents the average length of defined gesture g and is represented as Eq.(11) $L_{g} = \frac{\sum_{i = 1}^{m} L_{gi}}{m},$ (11) where L_gi denotes the i_th sample of defined gesture g and m denotes the total number of gesture samples.

2.3.2 Gesture segmentation

For gesture sequence input, define gesture sequence at moment t as O_t = {o₁, o₂, …, o_t} where o_i denotes discretized gesture data at moment i and o_i ∈ S. Calculate the similarity between O_t and global template Y_g of defined gestures based on GTDTW. It is defined as D_gt and showed as Eq.(12) $D_{gt} = {O_{t}, Y_{g}} .$ (12)

Find the gesture start point. The increment of D_gt between moment t and moment (t - 1) is defined as Δ_gt. It is showed as Eq.(13) $Δ_{gt} = D_{gt} - D_{g (t - 1)} .$ (13) When there is not a gesture start point, find the gesture start point according to observe the change of Δ_gt. The gesture start point is defined as g_start and showed as Eq.(14) $g_{start} = \underset{t}{argmax} (Δ_{gt}) .$ (14)

Find the gesture end point. After getting the start point of gesture g, the gesture end point is defined as g_end and g_end = t′ when Δ_gt′ = 0. After getting the gesture end point, continue to find the next start point so as to segment the whole gesture sequence.

The segmented gesture sequence can be got according to the gesture start and end point. The recognition process is combined with length characteristic. If the length of segmented gesture sequence meets Eq.(15), recognition result of segmented gesture sequence is gesture g. If not, recognition result of it is a no-gesture.

The gesture segmentation system proposed in this paper is forward spot continuous gesture segmentation system. It can output recognition result directly after segmentation to ensure the real-time performance. Combined with gesture length threshold, it can avoid the impact of no-gestures. Figure 4 illustrates how gesture recognition works based on GTDTW and Fig. 5 illustrates the gesture segmentation system.

Fig.4

How gesture recognition works based on GTDTW.

Fig.5

The gesture segmentation system.

3 Experiments and results

3.1 Experiment setup

The tool used in the experiment was MATLAB2014 software. For capturing gesture data, a wearable gloves device was produced which was produced by Beijing Noitom Company. The wearable glove was showed in Fig. 6. One glove consisted of nine sensor nodes in the finger part, one sensor node at the hand back and one sensor node at the hand wrist. Every sensor node was a 6-axis inertial sensor STM32-6050. Using the Axis Neuron Pro software provided by Noitom Company, it could automatically get the acceleration, angular velocity and quaternion data of the corresponding sensor node which could denote the curvature degree of the fingers and the direction of the palms. According to collecting the data of hand sensor nodes as features, we could describe the hand posture. The sample frequency of the sensor was 120Hz. In the experiments, thirty-three dimensional features were obtained at each moment according to the wearable glove device. There are a total of eleven sensor nodes on a glove. Every sensor nodes can get three dimensional features at each moment. In this work, nine gestures of right hand were defined for use based on Chinese Sign Language Manual. The sign meanings of nine gestures were greeting, thanking, eighty, staying up, walking, pasting, flight, you and catching that were showed in Fig. 7.

Fig.6

The wearable glove.

Fig.7

Description of different defined gestures.

3.2 Sample processing

We collected samples for both training the global templates of different defined gestures and evaluating the proposed method. Each defined gesture was collected a total of 120 samples, of which 80 as training samples, 40 as test samples and each sample could be denoted as Eq.(15) $X = (\begin{matrix} x_{1, 1} & x_{1, 2} & \dots & x_{1, t} \\ x_{2, 1} & x_{2, 2} & \dots & x_{2, t} \\ ⋮ & ⋮ & ⋱ & ⋮ \\ x_{n, 1} & x_{n, 2} & \dots & x_{n, t} \end{matrix}),$ (15) where x_i,j denoted the data of the i^th feature at moment j. K-means clustering algorithm was used to make samples discretized. In the experiments, the clusters number M of K-means was set to 15.The t in Eq.(15) is not a constant. Before discretized, the shape of both training samples, gesture sequence and segmented gesture sequence is same as Eq.(15). For a gesture or a gesture sequence, it may last t moments and we can obtain n features as a vector at each moment from data glove. Eq.(15) is a form of gesture data before discretized. It does not specifically represent a gesture or a gesture sequence.After discretized, each sample could be represented as Eq.(16) $X = (x_{1}; x_{2}, . . ., x_{t}),$ (16) where x_j denoted discretized gesture data at moment j. In other words, the gesture data at each moment became an integer which denoted the cluster that the gesture data belonged to.

3.3 Train global template and length threshold

Train global template. For each defined gesture, find the key frame and the sub-key frame according to the method proposed in Chapter 2.1 to form the global template. In the experiments, the global template of each defined gesture was showed in Table 1 and the weights of key frame and sub-key frame were showed in Table 2.

Train length threshold. For each defined gesture, calculate average length of the defined gesture samples as the length threshold L_g to represent the length characteristic. In the experiments, there were 80 samples for each defined gesture and the L_g was calculated as Eq.(17) $L_{g} = \frac{\sum_{i = 1}^{80} L_{gi}}{80},$ (17) where L_gi denotes i_th sample of defined gesture g. The length threshold of each defined gesture was showed in Table 1.

Table 1

The global template and length threshold of each defined gesture

Gesture ID		Greeting	Thanking	Eighty	Staying up	Walking	Pasting	Flight	You	Catching
Global template	Key frame	5,11,12	3,13	2,5,11,15	8,10	1,2,6	4,8	9	7,14	1,13
Sub-key frame	3	none	6	none	none	13,14	none	none	2,4,10
Length threshold		90	42	97	45	83	50	55	80	50

Table 2

The weights of key frame and sub-key frame

	Key frame	Sub-key frame
Weight	0.0667	0.0067

3.4 Isolated gesture recognition results

For isolated gesture recognition, the process of calculating similarity D_g between gesture sequences X and global templates Y_g was showed as Fig. 8. In Fig. 8, the states 1 and 2 were the key frame and state 3 was the sub-key frame. When states 1 and 2 appeared in gesture sequences X, the similarity increased corresponding weight 0.0667. The similarity increased 0.0067 when state 3 appeared and it did not increase when the others states appeared. In the experiments, test samples were used to evaluate the performance of proposed method and experiments were repeated five times to get the average as the results for ensuring the robustness. As showed in the experiments, the length of DTW template was usually 50 to 100 and the length of GTDTW template was less than 5. The time consumption was reduced to ensure the real-time performance. The result of isolated gesture recognition rate based the proposed method was showed in Table 3. When testing samples was 40, the comparison of time consumption and recognition rate of isolated gesture recognition based on different methods was showed in Table 4. The time consumption was the time taken to recognize 20 gestures. Experimental results showed that the performance of proposed method was better to other methods.

Fig.8

An illustration of the process of calculating similarity.

Table 3

The result of isolated gesture recognition rate based GTDTW

Gesture ID	Train samples	Test samples	Miss number	Recognition number	Recognition rate(%)
Greeting	80	40	4	36	90%
Thanking	80	40	0	40	100%
Eighty	80	40	0	40	100%
Staying up	80	40	2	38	95%
Walking	80	40	0	40	100%
Pasting	80	40	0	40	100%
Flight	80	40	0	40	100%
You	80	40	0	40	100%
Catching	80	40	0	40	100%
Average Recognition Rate(%)					98.3%

Table 4

The comparison of time consumption and recognition rate of isolated gesture recognition based on different methods

Method	Time consumption(s)	Average recognition rate(%)
HMM	1.85s	93.4%
DTW	7.00s	82.7%
Improved DTW [4]	2.83s	88.9%
Multi-templates DTW(3 templates) [6]	22.14s	91.1%
Multi-templates DTW(5 templates) [6]	36.62s	92.7%
GTDTW	0.17s	98.3%

3.5 Continuous gesture recognition results

For continuous gesture recognition, gesture segmentation was used to find gesture start and end points. In the experiments, 180 gestures were randomly made in a continuous gesture sequence with no-gestures between them. The sequence contained 20 valid gestures of each defined gesture. Part of segmentation process was shown in Fig. 9. Different colors were corresponding to different gestures.To evaluate the performance of temporal segmentation algorithms, recall rate and recognition rate were analyzed as quality criteria

Segmentation rate. The recall rate was that correct segmentation gesture number divided by total gestures number.

Recognition rate. The recognition rate was that correct recognition gesture number divided by total gestures number.

The proposed method was real-time. For vector discretization by K-Means clustering, each cluster center vector was recorded in the training process. When the test gesture sequence was input, it just needs to calculate distance between data vector at each moment and each cluster center vector. The time complexity of the processing is O (M) that M is the cluster numbers. It was constant complexity. For gesture segmentation, the similarity was calculated and compared between data vector after discretized and each global template. It was also constant complexity. In general, the time complexity of the total processing was constant complexity.

Fig.9

An illustration of the proposed method.

The process of experiments was repeated 5 times, calculating the average recall rate and recognition rate as results. The experimental results of continuous gesture recognition based on GTDTW were showed in Table 5. The experimental results for the two criteria demonstrated the effective performance of the gesture segmentation system. The length threshold was also proven to be useful in ensuring recognition performance of continuous gesture recognition that was showed in Fig. 10. The comparison of recognition rate of continuous gesture recognition based on different methods was showed in Table 6. The time consumption of gesture segmentation processing was showed in Table 7 and it just need 0.407s to segment 180 gestures. This is a nearly real-time processing for hand-robot interaction applications.

Fig.10

An illustration of the proposed method.

Table 5

The results of continuous gesture recognition based on GTDTW

Testing ID	Gestures number	Segmentation number	Segmentation rate(%)	Recognition number	Recognition rate(%)
1	180	180	100%	172	95.6%
2	180	178	98.9%	172	95.6%
3	180	178	98.9%	171	95.0%
4	180	180	100%	174	96.7%
5	180	179	99.4%	172	95.6%
Average	180	179	99.4%	172	95.6%

Table 6

Continuous gesture recognition rate based on different methods

Method	Recognition rate(%)
Forward accumulative HMM [28]	95.4%
Multi-size windows HMM [29]	89.6%
Improved DTW [30]	95.9%
Deep dynamic neural networks [15]	96.3%
Shot boundary detection [32]	93.0%
GTDTW gesture segmentation	95.6%

Table 7

The time consumption of gesture segmentation processing

Segment gestures number	Time consumption(s)
30	0.062s
60	0.129s
90	0.193s
120	0.268s
150	0.347s
180	0.407s

4 Conclusion

In this paper, the GTDTW algorithm is proposed for gesture recognition and a gesture segmentation system based on the proposed method is proposed for continuous gesture recognition. The wearable gloves are used to collect gesture data for gesture recognition system. Theory and experimentation indicate that GTDTW based on global templates can speed up continuous gesture recognition. For isolated gesture recognition, the proposed method improves recognition rate and recognition speed of isolated word gesture recognition.

There is some directions for future work. The size of experimental data set can be increased. It can also create more data sets to verify better robustness of the proposed method. For gesture segmentation, accumulation of errors may occur. It will lead to increase error rate in the second half of the gesture segmentation process. This problem requires a better way to solve.

Footnotes

Acknowledgments

Research was supported by National Natural Science Foundation of China (61372142, U1401252) and Fundamental Research Funds for the Central Universities SCUT (2017MS062). In addition, thanks to Baiyi Zhou for providing picture of hand gestures.

References

Simao

, Neto

and Gibaru

, Unsupervised gesture segmentation by motion detection of a real-time data stream, IEEE Transactions on Industrial Informatics 13 (2017). 473–481.

Chen

, Wang

, Deng

and Ji

, A Survey on Hand Gesture Recognition, 2013 International Conference on Computer Sciences and Applications 2013. pp. 313–316.

Bruno

, Mastrogiovanni

and Sgorbissa

, Wearable inertial sensors: Applications, challenges, and public test benches, IEEE Robotics & Automation Magazine 22 (2015). 116–124.

Zhang

and Qin

, DTW speech recognition algorithm of optimization template matching, World Automation Congress 2012 (2012). 1–4.

Choi

, Kim

and Kim

, A DTW gesture recognition system based on gesture orientation histogram, The 18th IEEE International Symposium on Consumer Electronics (2014). pp. 1–2.

Ruan

and Tian

, Dynamic gesture recognition based on improved DTW algorithm, 2015 IEEE International Conference on Mechatronics and Automation 2015. pp. 2134–2138.

Hussain

and Rashid

, User independent hand gesture recognition by accelerated DTW, 2012 International Conference on Informatics, Electronics & Vision 2012. pp.1033–1037.

Abdulla

W.H.

, Chow

and Sin

, Cross-words reference template for DTW-based speech recognition systems, TENCON 2003 Conference on Convergent Technologies for Asia-Pacific Region 2003. pp. 1476–1579.

Jambhale

and Khaparde

, Gesture recognition using DTW & piecewise DTW, International Conference on Electronics & Communication Systems 2014. pp. 1–5.

10.

Eickeler

, Kosmala

and Rigoll

, Hidden Markov model based continuous online gesture recognition, Proceedings Fourteenth International Conference on Pattern Recognition 1998. pp.1206–1208.

11.

Shrivastava

, A hidden Markov model based dynamic hand gesture recognition system using OpenCV, 2013 3rd IEEE International Advance Computing Conference 2013. pp. 947–950.

12.

Yang

and Sarkar

, Gesture Recognition using Hidden Markov Models from Fragmented Observations, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2006. pp. 766–773.

13.

Shin

and Sung

, Dynamic hand gesture recognition for wearable devices with low complexity recurrent neural networks, 2016 IEEE International Symposium on Circuits and Systems (2016). pp. 2274–2277.

14.

Zeng

, Xiao

and Li

, A K-NN and Sparse Representation Based Method for Gesture Recognition, 2013 IEEE 10th International Conference on High Performance Computing and Communications & 2013 IEEE International Conference on Embedded and Ubiquitous Computing 2013. pp. 2325–2329.

15.

Neto

, Pereira

, Pires

J.N.

and Moreira

A.P.

, Real-time and continuous hand gesture spotting: An approach based on artificial neural networks, 2013 IEEE International Conference on Robotics and Automation 2013. pp. 178–183.

16.

Edel

and Koppe

, Binarized-BLSTM-RNN based Human Activity Recognition, 2016 International Conference on Indoor Positioning and Indoor Navigation 2016. pp. 1–7.

17.

, Pigou

, Kindermans

, Le

, Shao

, Dambre

and Odobez

, Deep dynamic neural networks for multimodal gesture segmentation and recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence 38 (2016). 1583–1597.

18.

Song

, Gu

, Wang

, Liu

and Li

, A Kinect based gesture recognition algorithm using GMM and HMM, 2013 6th International Conference on Biomedical Engineering and Informatics 2013. pp. 750–754.

19.

Zhang

, Wang

and Ma

, Improvement of Dynamic Hand Gesture Recognition Based on HMM Algorithm, 2016 International Conference on Information System and Artificial Intelligence 2016. pp. 401–406.

20.

Delgado-Mata

and Cosio

, HMM and NN for Gesture Recognition, IEEE Electronics, Robotics and Automotive Mechanics Conference (2010). pp. 56–61.

21.

, Li-Shuang

, Lei

and Xue-Bo

, Dynamic hand gesture recognition using HMM-BPNN model, 2016 IEEE International Conference on Real-time Computing and Robotics 2016. pp. 422–426.

22.

Liao

, Li

, Sun

, Jiang

, Kong

, Jiang

and Ju

, Static hand gesture segmentation: Comparison and selection of existing methods, 2017 32nd Youth Academic Annual Conference of Chinese Association of Automation 2017. pp. 889–894.

23.

Simao

M.A.

, Neto

and Gibaru

, Unsupervised gesture segmentation of a real-time data stream in MATLAB, IECON 2016 - 42nd Annual Conference of the IEEE Industrial Electronics Society 2016. pp. 809–814.

24.

, Wang

, Zeng

, Chen

and Liu

, Depth and RGB image alignment for hand gesture segmentation using Kinect, 2013 International Conference on Machine Learning and Cybernetics 2013. pp. 913–919.

25.

Yao

and Li

, Hand Gesture Segmentation in Uncontrolled Environments with Partition Matrix and a Spotting Scheme Based on Hidden Conditional Random Fields, 2013 2nd IAPR Asian Conference on Pattern Recognition 2013. pp. 842–846.

26.

Shin

and Sung

, Dynamic hand gesture recognition for wearable devices with low complexity recurrent neural networks, 2016 IEEE International Symposium on Circuits and Systems (2016). pp. 2274–2277.

27.

Deng

J.W.

and Tsui

H.T.

, An HMM-based approach for gesture segmentation and recognition, Proceedings 15th International Conference on Pattern Recognition ICPR-2000 2000. pp. 679–682.

28.

Song

and Kim

, Simultaneous Gesture Segmentation and Recognition based on Forward Spotting Accumulative HMMs, 18th International Conference on Pattern Recognition 2006. pp. 1231–1235.

29.

Chavarria

H.V.

, Escalante

H.J.

and Sucar

L.E.

, Simultaneous segmentation and recognition of hand gestures for human-robot interaction, 2013 16th International Conference on Advanced Robotics 2013. pp.1–6.

30.

Kuzmanic

and Zanchi

, Hand shape classification using DTW and LCSS as similarity measures for vision-based gesture recognition system, EUROCON 2007 - The International Conference on “Computer as a Tool” 2007. pp. 264–269.

31.

and Greenspan

, Segmentation and Recognition of Continuous Gestures, 2007 IEEE International Conference on Image Processing 2007. pp. 365–368.

32.

, Zhu

and Shao

, One shot learning gesture recognition from RGBD images, 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops 2012. pp. 7–12.

33.

, Nguyen

, Tran

, Nguyen

and Nguyen

, Temporal gesture segmentation for recognition, 2013 International Conference on Computing, Management and Telecommunications 2013. pp. 369–373.

34.

Elmezain

, Al-Hamadi

and Michaelis

, Hand trajectory-based gesture spotting and recognition using HMM, 2009 16th IEEE International Conference on Image Processing 2009. pp. 3577–3580.

35.

Roy

, Mahbub

Md , Rahman

, Imtiaz

Md , and Ahad

A.R.

, Temporal segmentation of gestures using gradient orientation of depth images, 2013 International Conference on Informatics, Electronics and Vision 2013. pp. 1–5.

36.

Beh

, Han

and Ko

, Rule-based trajectory segmentation for modeling hand motion trajectory, Pattern Recognition 47 (2014). 1586–1601.