Yoga Posture Recognition for Self-training

Chen, Hua-Tsung; He, Yu-Zhen; Hsu, Chun-Chieh; Chou, Chien-Li; Lee, Suh-Yin; Lin, Bor‐Shyh

doi:10.1007/978-3-319-04114-8_42

Cited by 53 publications

(29 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Multiple Linear Regression is used to estimate the values used in Choquet integral [11]. For observation variable x1, x2, ..., xn which is also called predictive attribute, and y as objective attribute, with l sets of data training, described as a matrix below [9]: Linear pattern model has also already been developed in Inverse Function Neural Network using linear function as activation function [20] and has been applied in image object pattern recognition for static images [21].…”

Section: Feature Modelling Using Choquet Integralmentioning

confidence: 99%

“…For body posture, we use silhouettes from background subtraction using Kinect, and then the contour can be extracted using OpenCV showed in Figure 11. We then try to match pairs of silhouettes and also contours using SURF [11] in simulated movement which also include various scales and illuminations. Then we count the correct match compares to the wrong match to calculate the true positive and the false positive rates.…”

Section: Combination Of Face and Posture Features For Tracking Of Movingmentioning

confidence: 99%

“…Silhouettes matching using SURF [11] only cannot satisfy accurate object recognition since they results in high false positives. Then the high false positive rate can be lowered using contour of the silhouettes which also result in low true positives.…”

Section: Combination Of Face and Posture Features For Tracking Of Movingmentioning

confidence: 99%

“…This paper continues the work in combination of the face and body posture movements by utilization of Opencv Dlib face landmarks [24,25], Kinect [5] and image contouring in Opencv, and it is compared with SURF methods [11].…”

Section: Introductionmentioning

confidence: 99%

“…Structural hierarchical and relational methods have been done for high level abstraction of image objects modelling, but still have problems for low level implementation [7][8]. Many researchers have developed various human activities and behavior recognition, such as walking, sitting, bending, and some sport activities, and less work of person identification that does the activities [11]. Gait identification has been done by [3], but not yet using the whole body combined with face in changing poses for personal identification.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Combination of Face and Posture Features for Tracking of Moving Human Visual Characteristics

Andriana¹,

Prihatmanto²,

Hidayat³

et al. 2017

ijeei

View full text Add to dashboard Cite

Understanding images by recognizing its objects is still a challenging task. Tracking of moving human and recognition have been developed by researchers but not yet shows enough information needed for recognition. Initially, a tracking process of an object starts with detection and recognition of the object in a static pose and position, and then continues in movement in different poses. Available moving human recognition methods still has error in classification and need a huge amount of examples which may still be incomplete. Human face and body posture characteristics such as size of the eyes, nose, mouth, or fat or thin bodies, are important visual features in different poses for personal identification to increase accuracy of human recognition system, and it is still rare in researches. This paper attempts to describe visual features that best known for human, but hard to be recognized by machines. Curve fitting approaches to face and body posture features are also introduced to capture exact patterns of the features. Body postures are also preprocessed with a Kinect depth camera, and also compared to popular and recent methods of visual object recognition. Finally, we demonstrate our method can be useful for visual object classification. Probabilities of personal identification can be increased by using different poses and characteristics of smaller detail features through body postures and face areas. More detail features will richen comparison data samples for higher recognition accuracy.

show abstract

Section: Feature Modelling Using Choquet Integralmentioning

confidence: 99%

Section: Combination Of Face and Posture Features For Tracking Of Movingmentioning

confidence: 99%

Section: Combination Of Face and Posture Features For Tracking Of Movingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%