e-Yoga Prescription Designed for Computer Users using e-Yoga

Environment for Posture

A Study on Information Technology (IT) Professionals to Help Them Improve

Their Spine Health

M. Shivarama Reddy

, Suhasini A. Venkatramana

, B. R. Ramji

and Nishreyas M. V.

Department of Sports and Cultural Activities, BMS College of Engineering, Bangalore, India

Department of Industrial Engineering and Management, BMS College of Engineering, Bangalore, India

Department of Computer Science, Jain University, Bangalore, India

Keywords: e-Yoga, Motion Analysis, Video Analysis, Virtual Reality, Yoga Avatar, Yoga Tools, Yoga Prescription,

Yogic Cure.

Abstract: The paper presents the next generation health protection techniques using postures and exercises of Yoga

specially designed for IT professionals suffering with sever Spine related health issues. On a short survey

conducted on major health issue concerning the Information Technology - IT professionals here in

Bangalore, India. The result shows Posture and Spine related issues made the work delay and employee

well-being is disturbed at higher levels dragging the employees into chronic pain and depression producing

less creativity. We formulated a general health check up to understand the nature of Yoga needed to help the

employees or users to improve their Gait and posture. Using Video Analysis techniques for simple Yoga

and motion analysis techniques for advanced yoga, we have designed a prototype where the user can reap

benefits of Yoga by creating a Virtual Yoga Environment at the work place or wherever the employee uses

the computer for a long time. The environment alerts the user at regular intervals depending on user Yoga

Prescription that is generated by the environment at the time of user requirement, inputs and accessibility.

The alerts for healthy users and young IT professionals will be blinking their eyes very often, standing up

from the work chair, stretch, a standing twist and some basic neck movements again depending on the body

size of the user. However performing these simple Yoga exercises on the whole in a day would be

approximately 5 to 7 minutes for a6 to 8 hours Computer user that would promote better health.

1 INTRODUCTION

In this paper we have made an effort to design an e-

yoga tutor for the health issues of computer users

especially in the Information technology - IT

industry for whom we have designed a real time

Yoga prescription to improve the chronic health

conditions like of those suffering from Spine related,

posture issues and other psychological problems like

depression. It not only helps in treating their health

conditions alternatively it helps users to improve

concentration levels and beat stress levels at work.

The primary work of our research was to

investigate the health issues among the IT users.

Based on the age of the users and duration of work

hours the employee used a computer, the results

were tabulated. Using this data we categorised the

health issues for which suitable Yoga treatment is

designed. During the work hours we tried to imbibe

the yoga exercises to be performed and hence the

need of an effective intelligence system that

interfaced human and computer was essential. This

is well explained in the further sections.

2 GENERAL HEALTH CHECK

The general health check details consist of basic

details mentioned in Table 1. These details act as

input for yoga prescription. Yoga prescription is

letting the user know which Yoga asana or exercise

of yoga is suitable for the cure if any ailment present

or for general well-being. The health check for IT

professionals under different categories of computer

usage are been categorised making use of the table

for general Yoga prescription. These basic details

Reddy, M., Venkatramana, S., Ramji, B. and V., N..

e-Yoga Prescription Designed for Computer Users using e-Yoga Environment for Posture - A Study on Information Technology (IT) Professionals to Help Them Improve Their Spine Health.

In Proceedings of the 3rd International Congress on Sport Sciences Research and Technology Support (icSPORTS 2015), pages 48-52

ISBN: 978-989-758-159-5

are used for creating a Yoga Avatar that is used for

interfacing between real and virtual environment.

Table 1: Preliminary Database of general health check for

yoga prescription.

Age Group (20 – 25, 25 – 30, 30-45)

sex (Male/Female)

No of hours Computer used per

day

(2-5, 6-9, >9)

No of hours sleep (<4, 5-7, 8-10, >10)

Height In Inches

weight In Kilo Grams

Eye Sight (Normal/ Prescription)

Undergoing Treatment Details

surgery undergone If any

No of years’ Experience (0-25)

Current Domain Experience (0-25)

Need any help Health Related

3 YOGA AVATAR

The size of the human is taken into consideration for

creating a human avatar virtually hence we refer

here it as Yoga Avatar. The need of Yoga Avatar is

very much essential for yoga prescription for the

following:

 Yoga Asana should always be performed under

guidance. The steps beginning from inhaling and

exhaling during the exercise to proper level of

stretching and bending plays a vital role for

which the Yoga Avatar is mandatory to

determine the level of stretch. Not all have a

unique body size that Yoga can be performed at

the same level.

 It acts like a watermark animated picture that

does the asana virtually on screen depending

upon the human size where even a person

weighing 120 Kilos can twist to a certain degree

that the avatar knows while a person weighing 50

kilos can do it more. This helps that helps the

user perform along the Yoga Avatar do a better

Yoga Asana.

Figure 1: Yoga Avatar.

 The human body size changes constantly and it

would be tough to measure the sizes every day or

very often to update into the software to change

the yoga avatar. Hence we have created an

option of live yoga Avatar that captures your

present moment and suggests you the degree of

twist or turn or bending for the asana.

4 VIRTUAL YOGA

ARCHITECTURE

Analysing the motion of the human body includes

the extraction of the low-level features like body

part segregation, body parts detection and

identification, and the recovery of 3D structure from

the 2D projections in an image sequence. To trackan

individual moving using a single camera that is

preferably a webcam or a mobile camera involves

applying visual features to detect the presence of

human directly with considering only the geometric

structure of the body parts.

Motion information such as change in position

can be captured constantly to match the needed

change in the virtual environment, incorporating

with intensity values that can be employed to

establish matching between consecutive frames. The

correspondence between successive frames from real

world and virtual world are solved throughout using

the image sequence.

The SIFT (Scale Invariant Feature Transform)

algorithm can be used for comparing the two videos

the reference or virtual video and the actual video.

This technique was originally devised for object

recognition. The features like good variance,

rotation and illumination SIFT descriptor was

commonly used in image matching. (Daixian, 2010)

Applications of SIFT include object recognition,

robotic mapping and navigation, image stitching, 3D

modelling, gesture recognition, video tracking, and

match moving. Using SIFT as base for Virtual Yoga

Architecture SURF (Speeded Up Robust Feature)

Algorithm the advanced version can be looked upon.

The major advancement is 3D reconstruction.

SIFT and SURF algorithms employ slightly

different ways of detecting features. SIFT builds an

image pyramids, filtering each layer with Gaussians

of increasing sigma values and taking the difference.

On the other hand, SURF creates a “stack” without

2:1 down sampling for higher levels in the pyramid

resulting in images of the same resolution. Due to

the use of integral images, SURF filters the stack

using a box filter approximation of second-order

Gaussian partial derivatives, since integral images

e-Yoga Prescription Designed for Computer Users using e-Yoga Environment for Posture - A Study on Information Technology (IT)

Professionals to Help Them Improve Their Spine Health

allow the computation of rectangular box filters in

near constant time. (Juan and Gwun, 2009).

Figure 2: This caption has one line so it is centred.

Figure 3: Video frame after thresholding.

Figure 4: Comparison lines between frames of two videos

after SURF algorithm.

Figure 5: Implementation using SURF (Patil et al., 2011).

4.1 Motion Replication

Motion Replication ensures the understanding of the

user’s motion by matching the user’s posture with

the Yoga Avatar posture. This replication is done by

analysing key nodes of body parts as shown in the

figure 6. This recognition of body parts that are not

replicating the corresponding ones shows the

difference in posture and need to complete the

sequence as mentioned in the virtual environment.

The methodology for determining the similarity of

two postures is by comparing the joint angles of

fifteen joints, including head, shoulders, elbows,

wrists, chest, pelvis, hips, knees, and ankles, and by

calculating the distance of end points required in

each Yoga posture. (Zhiqiang et al., 2011).

Figure 6: KEY nodes of body parts to Understand Motion.

icSPORTS 2015 - International Congress on Sport Sciences Research and Technology Support

4.2 Yoga Tools

Yoga tools are extended products that are interface

suits that connect the real human with the Virtual

Yoga environment. In Video analysis the image is

captured of the user and matched with its yoga

Avatar animation in which the computer or the

environment cannot understand the intensity and

pressure required for the posture. With the help of

yoga tools that use hardware encompassed of

sensors and motors to measure the real environment

vibrations, pressure, and velocity of the posture to

measure the accuracy and use of the posture.

Devices with sensors can measure Roll-Pitch-

Yaw orientations, which is used to detect the

orientation of human body part. The tactor is

attached on four human limbs and can generate the

vibration on human skin to provide the haptic sense

to user. (Zhiqiang Luo et al., 2011)

4.3 e-Yoga Development

The e-Yoga Environment can be run on a computer,

Mobile Phone, Tablet. We have created an e- yoga

version with General Health Check database that

collects the user basic data to generate an e-Yoga

Prescription that helps user perform basic yoga

asana’s. The technical requirements are that it

should be installed in the hard drive that gives

regular alerts at intervals asking the user to perform

the Yoga.

4.4 Side Effects

It is difficult to change the mind suddenly from a

focused state, when the software from the system

alerts to perform a regular bit of Yoga it could be a

little burden and distracting sometimes. However the

user can mute the Virtual Yoga off. The user can be

aware and can be prepared to perform yoga even

before the environment alerts so that it will help the

user in a long run even without a virtual yoga tutor

or environment.

5 CONCLUSIONS

The e-Yoga environment is a next generation

intelligence that is designed to meet the requirement

of a traditional Yoga Guru (tutor) who helps an

individual maintain a good health, body and mind.

This environment creates alerts at regular intervals

on the device used be it on a computer, mobile

phone, tablet or a PlayStation. The alert depends on

users Yoga Prescription that is generated by the E-

yoga environment at the time of user requirement,

inputs and accessibility. The alerts for users with no

health issues and young IT professionals the yoga

prescription is simple like blinking their eyes very

often, standing up from the work chair and stretch, a

standing twist and some basic neck movements

which again is depending on the body size of the

user. However performing these simple Yoga

exercises on the whole in a day would be

approximately be 5 to 7 minutes for a 6 to 8 hours

Computer user. Hence it would not be a tedious

process to keep oneself fit and healthy and prevent

from further complications. The application under

development can be designed for mobile, Tablet and

PlayStation users equally to maintain their gait and

posture. Our mission is to promote a healthy society.

ACKNOWLEDGEMENTS

We take immense pleasure to thank Mrs.

Siddalakshmi D, Director of Ierosun Technologies

Pvt Ltd for her valuable inputs regarding Video

analysis and Motion analysis for Virtual

yogaArchietecture. It was an honour to work with

Mr. GVenkata Chalapathi, Physical Education

Director, Government First Grade College,

Harlapura, Koppala, Karnataka, India who

constantly guided us with posture improvement

methods. We also take this opportunity to thank

Mrs. Sheela, Yoga tutor for her Yoga tutorials and

survey of users and respective Yogic cure that has

made us possible to incorporate appropriate data to

the Virtual Yoga Architecture. Lastly, we thank

BMS college of Engineering for constantly

encouraging research activities and supporting

research endeavours.

REFERENCES

Emre Sariyildiz, Kouhei Ohnishi, 2014. A Guide to design

Disturbance Observer based Motion Control Systems.

The 2014 International Power Electronics

Conference.

Yu Kinoshita, Akinori Sasaki, Hiroshi Hashimoto and

Chihiro Ishii, 2007.Analysis and Evaluation of Hand

Motion in Use of Tools. IEEE.

J. Foody, D. Kelly, D. Kumar, D. Fitzgerald, T. Ward, B.

Caulfield, C. Markham, 2006.A Prototype Sourceless

Kinematic-Feedback Based Video Game for

Movement Based Exercise. Proceedings of the

EMBS Annual International Conference New York

City, USA, IEEE.

e-Yoga Prescription Designed for Computer Users using e-Yoga Environment for Posture - A Study on Information Technology (IT)

Professionals to Help Them Improve Their Spine Health

Zhiqiang Luo, Weiting Yang, Zhong Qiang Ding, Lili

Liu1, I-Ming Chen, Song Huat Yeo, Keck Voon Ling,

Henry Been-Lirn Duh, 2011. “Left Arm

Up!”Interactive Yoga Training in Virtual

Environment. IEEE Virtual Reality 201.

Siddharth Patil, Amey Pawar, Aditya Peshave, Aamir N.

Ansari, Arundhati Navada, 2011. Yoga Tutor,

Visualization and Analysis Using SURF Algorithm.

IEEE Control and System Graduate Research

Colloquium.

Zhu Daixian, SIFT algorithm analysis and optimization-

April 2010; Image Analysis and Signal Processing

(IASP).

GustiBagus Baskara Nugraha, 2014, Video Analysis Tools

For Cloud-based Motion Detection, IEEE.

Luo Juan and Oubong Gwun, August 2009. A comparison

between SIFT, PCA-SIFT and SURF, CSC Journals,

Kuala Lumpur, Malaysia,

‘Human Motion Analysis – A Review’ – Nonrigid and

Articulated Motion Workshop, 1997. Proceedings,

IEEE.

Alexander Kuhn, Tobias Senst, Ivo Keller, Thomas

Sikora, Holger The isel, 2012. A Lagrangian

Framework for Video Analytics, IEEE.

Nils Jäger, Stuart Moran, Holger Schnädelbach, 2014.

Using Adaptive Architecture to Support

YogaPractices.

The Third IEEE International Workshop on Social

Implications of Pervasive Computing.

Salsabeel F. M. Alfalah, Jannat Falah, Warren Chan,

Tasneem Alfalah, Soheeb Khan, David K. Harrison,

Vassilis Charissis, 2014. Gait Analysis Data

Visualisation in Virtual Environment. Science and

Information Conference.

Zhou Kaii, Tian Fengi, Hao Guo, Ren Zhong, 2014. A

Clustering Compression Method for 3D Human

Motion Capture Data. The 9th International

Conference on Computer Science & Education

(ICCSE 2014) August 22-24, 2014. Vancouver,

Canada.

Rahul R Pitale, Kapil D Tajane, Prof.(Dr).Jayant S.

Umale, 2014. Detection of Non-Linear Characteristics

of HRV patterns for different yoga posture.

International Conference for Convergence of

Technology.

Zulj, S., Sekata, G., Dzaja, D., Celic, L., & Magjarevic, R.

(2015, January).Virtual reality system for Assisted

Exercising Using WBAN. In 6th European Conference

of the International Federation for Medical and

Biological Engineering. Springer International

publishing.

icSPORTS 2015 - International Congress on Sport Sciences Research and Technology Support