Hidden Markov Model on a unit hypersphere space for gesture trajectory recognition

Beh, Jounghoon; Han, David K.; Durasiwami, Ramani; Ko, Hanseok

doi:10.1016/j.patrec.2013.10.007

Cited by 36 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…True values from the confusion matrix are used to determine the correct matching features from the hand gesture model. The comparison of all the classes of hand gesture data is shown in Table 9 using HMM with SSA and Viterbi approach for determining the recognition rate and error rate calculation by using Equations (17) and (18). In this table, three basic positions like starting, middle, and ending position of the hand gesture pattern have been taken.…”

Section: Classification and Accuracy Resultsmentioning

confidence: 99%

“…Hence, a normalized longest common subsequence is used to achieve better results. Beh et al have adopted a mixture of von Mises‐Fisher with HMM for gesture trajectory. The results were analyzed with publicly available hand gesture data, University of Central Florida (UCF) Kinect, and InteractPlay in terms of recognition rate and scalability.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A probabilistic model for state sequence analysis in hidden Markov model for hand gesture recognition

Sagayam

Hemanth

2018

Computational Intelligence

View full text Add to dashboard Cite

The role of gesture recognition is significant in areas like human‐computer interaction, sign language, virtual reality, machine vision, etc. Among various gestures of the human body, hand gestures play a major role to communicate nonverbally with the computer. As the hand gesture is a continuous pattern with respect to time, the hidden Markov model (HMM) is found to be the most suitable pattern recognition tool, which can be modeled using the hand gesture parameters. The HMM considers the speeded up robust feature features of hand gesture and uses them to train and test the system. Conventionally, the Viterbi algorithm has been used for training process in HMM by discovering the shortest decoded path in the state diagram. The recursiveness of the Viterbi algorithm leads to computational complexity during the execution process. In order to reduce the complexity, the state sequence analysis approach is proposed for training the hand gesture model, which provides a better recognition rate and accuracy than that of the Viterbi algorithm. The performance of the proposed approach is explored in the context of pattern recognition with the Cambridge hand gesture data set.

show abstract

Section: Classification and Accuracy Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A probabilistic model for state sequence analysis in hidden Markov model for hand gesture recognition

Sagayam

Hemanth

2018

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…Pisharady & Saerbeck, 2015 [13] reviewed conventional hand-gesture recognition using RGB cameras as well as recognition using RGB-D sensors. Pisharady & Saerbeck, 2015 [13] classified the techniques used for dynamic hand gesture recognition as: (a) Hidde Markov Models (e.g., [16]) and other statistical methods (e.g., [17]); (b) Artificial Neural Networks (e.g., [18]) and other learning based methods (e.g., [19]); (c) Eigenspacebased methods (e.g, [20]); (d) Curve fitting [21]; and (e) Dynamic programming [22]/Dynamic time warping (e.g., [23]). Depth sensors have already been used in computer vision for many years both commercial and non-commercial (e.g., [24]).…”

Section: Gesture Interactionmentioning

confidence: 99%

Using a Serious Game to Assess Spatial Memory in Children and Adults

Loachamín-Valencia

Juan

Méndez-López

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Short-term spatial memory has traditionally been assessed using visual stimuli, but not auditory stimuli. In this paper, we design and test a serious game with auditory stimuli for assessing short-term spatial memory. The interaction is achieved by gestures (by raising your arms). The auditory stimuli are emitted by smart devices placed at different locations. A total of 70 participants (32 children and 38 adults) took part in the study. The outcomes obtained with our game were compared with traditional methods. The results indicated that the outcomes in the game for the adults were significantly greater than those obtained by the children. This result is consistent with the assumption that the ability of humans increases continuously during maturation. Correlations were found between our game and traditional methods, suggesting its validity for assessing spatial memory. The results indicate that both groups easily learn how to perform the task and are good at recalling the locations of sounds emitted from different positions. With regard to satisfaction with our game, the mean scores of the children were higher for nearly all of the questions. The mean scores for all of the questions, except one, were greater than 4 on a scale from 1 to 5. These results show the satisfaction of the participants with our game. The results suggest that our game promotes engagement and allows the assessment of spatial memory in an ecological way.

show abstract

“…Currently, the most influential methods for vision-based hand trajectory recognition include hidden Markov model (HMM), dynamic time warping (DTW), template matching, and traditional machine-learning algorithms. Jounghoon Beh et al proposed a gesture trajectory recognition method combining the mixture of von Mises-Fisher (MvMF) probability density function with HMM [5], which mitigates the problem caused by the change of trajectory in time and space. However, the method needs a long training time.…”

Section: Introductionmentioning

confidence: 99%