Final-smartmirror-1.docx

AN AUGMENTED REALITY SYSTEM USING SMART MIRROR TECHNOLOGY WITH DECISION SUPPORT IN FASHION OUTFIT SELECTION.

A Thesis Proposal Presented to the Faculty of the Information and Communications Technology Program STI College Caloocan

In Partial Fulfilment of the Requirements for the Degree Bachelor of Science in Computer Science

Jinky B. Fuentes Renz Adriane E. Ocampo Ray Clark D. Sulit Joan G. Villanueva

March 2019

ENDORSEMENT FORM FOR ORAL DEFENSE

TITLE OF RESEARCH:

“Mirror Tech”: An Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection.

NAME OF PROPONENTS:

Jinky B. Fuentes Renz Adriane E. Ocampo Ray Clark D. Sulit Joan G. Villanueva

In Partial Fulfilment of the Requirements for the degree Bachelor of Science in Computer Science has been examined and is recommended for Oral Defense.

ENDORSED BY:

Engr. Bernard A. Mendiola Thesis Adviser APPROVED FOR FINAL DEFENSE:

Mrs. Marilou F. Bartolome Thesis Coordinator

NOTED BY:

Prof. Julius P. Claour ICT Program Head

March 2019

STI College Caloocan

ii

APPROVAL SHEET This thesis titled: “Mirror Tech” An Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection. Prepared and submitted by Jinky B. Fuentes; Renz Adriane E. Ocampo; Ray Clark D. Sulit; Joan G. Villanueva, in partial fulfilment of the requirements for the degree of Bachelor of Science in BSCS, has been examined and is recommended for acceptance an approval. Engr. Bernard A. Mendiola Thesis Adviser

Accepted and approved by the Thesis Review Panel in partial fulfilment of the requirements for the degree of Bachelor of Science in Computer Science

Dennis S. Nava Panel Member

Engr. Marcelo Varona Panel Member

Prof. Julius p. Claour Lead Panelist

APPROVED:

Ms. Marilou F. Bartolome Thesis Coordinator

Prof. Julius P. Claour ICT Program Head

March 2019

STI College Caloocan

iii

ACKNOWLEDGEMENTS

The developers would like to thank the following:

Thesis Coordinator, Prof. Marilou F. Bartolome, who agreed to supervise this thesis. Without her great guidance and dedication, we would not have come this far and could have not accomplished our thesis paper. We would like to thank them for her patience and support throughout the process and helped us whenever we needed them over the past year.

We would like to express our gratitude to our beloved and respected acting thesis adviser Engr. Bernard A. Mendiola for his support and blessing to accomplish this project.

To our Family and friends, we would also like to express our gratefulness for their endless moral support and inspiring to finish this project.

Lastly, we would like to thank STI College Caloocan for giving us opportunity to complete our Bachelor of Science degree in Computer Science.

STI College Caloocan

iv

ABSTRACT Title of research: “Mirror Tech” An Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection. Researchers:

Jinky B. Fuentes Renz Adriane E. Ocampo Ray Clark D. Sulit Joan G. Villanueva

Degree:

Bachelor of Science in Computer Science

Date of Completion: March 2019 Key words:

Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection.

The advent of technology results to a great impact as it improves various phases in the human life. It enables man to do things for greater than ordinary ability and make daily work more efficient and a lot easier. Fashion refers to the popular style or practice, especially in clothing, footwear, accessories, makeup, hairstyle and body. Fashion is adistinctive and often a constant trend in style in which a person dress. It is the prevailing style in behavior and the newest creations which man observe and fellow to be "In". In today's modern generation, everyone enjoys shopping for daily goods and clothing. However, there is a small drawback in shopping at department stores like the hassles on queues, unavailability of fitting rooms, and being unable to make decisions on selected clothes or items. The researchers faced challenges in developing a fashion outfit selection in augmented reality platform. Using a sensor, the system will detect the user's skin tone. Through the decision support system, it will suggest possible fashion outfit for the upper and lower body which will be displayed in a smart mirror. It is a real-time fashion outfit system consisting multiple tasks like detecting and resizing input dress images. It used 3D fashion outfit design for the positioning and rotation of dress. The researchers were motivated to enable users to be in style and wear fashionable cloths suitable to their skin tone so they would look presentable and nice. Understanding the developed "Mirror Tech": An Augmented Reality System using Smart Mirror Technology guides and assists a person's need for attires which complements his/her skin tone. The system will produceprominently credible and well-fitted fashion outfit result.

STI College Caloocan

v

TABLE OF CONTENTS Page Title Page

i

Endorsement Form for Proposal Defense

ii

Approval Sheet

iii

Acknowledgements

iv

Abstract

v

Table of Contents

vi

List of Tables

vii

List of Figures

Viii

List of Appendices

ix

Introduction

1

Background of the problem

3

Overview of the current state of technology

4

Objectives of the study

5

Scope and limitations of the study

6

Literature Review

8

Review of related literature, studies or systems

8

Synthesis

14

“Mirror Tech” An Augmented Reality System using Smart Mirror

15

Technology with Decision Support in Fashion Outfit Selection. Results and Discussions

21

Conclusions and Recommendations

23

References Appendices

STI College Caloocan

vi

LIST OF TABLES Table

Page 1 Survey Result

STI College Caloocan

22

vii

LIST OF FIGURES Figure

Page 1 Prototype Model (MirrorTech)

15

2 (MirrorTech) Homepage

17

3 The scenario of using dynamic fitting room

18

4 Body index and gesture.

18

5 Scanner skin tone

19

6 Main function.

19

STI College Caloocan

viii

LIST OF APPENDICES Appendix

Page A. Gantt chart of activities B. Actual Thesis Expenses C. User’s Manual D. Curriculum Vitae of Researchers

STI College Caloocan

ix

INTRODUCTION

Augmented reality (AR) displays in a general sense, within the context of a reality - virtuality (RV) continuum, encompassing a large class of `mixed reality' (MR) displays, which also includes augmented virtuality (AV). MR displays are defined by means of seven examples of existing display concepts in which real objects and virtual objects are juxtaposed. Essential factors which distinguish different MR display systems from each other are presented, first by means of a table in which the nature of the underlying scene, how it is viewed, and the observer's reference to it are compared, and then by means of a three-dimensional taxonomic framework comprising: extent of world knowledge, reproduction fidelity, and extent of presence metaphor. A principal objective of the taxonomy is to clarify terminology issues and to provide a framework for classifying research across different disciplines. Fashion is a popular style or practice, especially in clothing, footwear, accessories, makeup, hairstyle and body. Fashion is a distinctive and often constant trend in the style in which a person dress. It is the prevailing styles in behavior and the newest creations.

Because the more technical term costume is regularly linked to the term "fashion", the use of the former has been relegated to special senses like fancy dress or masquerade wear, while "fashion" generally means clothing, including the study of it. Although aspects of fashion can be feminine or masculine. The use of technology is increasing day by day, we all depend on technology, and we use various technologies to accomplish specific tasks in our lives. Today we have various emerging technologies that impact our lives in different ways. Technology is being implemented in almost every aspect of our lives. We all know that technology helps us to do things far greater than our ability and resources. The technology makes us do easier to get more things in our daily routine. Many individuals create their own technology to expand the idea of technology so it can produce different technology.

STI College Caloocan

1

Studying Decision Support Systems (DSS), a specific class of computerized information system that supports business and organizational decision-making activities. A properly designed Decision Support System is an interactive software-based system intended to help decision makers compile useful information from raw data, documents, personal knowledge, and/or business models to identify and solve problems and make decisions. Smart mirrors are straight from science fiction. They’re part of an optimistic vision of the future that imagines a world where screens and data are everywhere, ready to feed you whatever information you need at a moment’s notice.

STI College Caloocan

2

Background of the problem Sometimes hear people talk about how blue is "their" color, or how they look awful in peach. What causes people to have flattering and clashing colors for their skin tone, and how do you determine which colors are best for you? We'll tell you how in just a moment, but try to remember that these rules aren't set in stone. If you love a color that isn't the best for you, that doesn't mean you have to stop wearing it! Consider using it as an accent color with a belt, scarf, shoes, or handbag instead of wearing it head-to-toe. How to Determine Skin Tone? Before you can choose your ideal clothing colors, you'll need to determine your skin tone. Skin tones vary from warm (yellow/green-based) to cool (pink/blue-based), and some people have a neutral skin tone that works well with both warm and cool tones. Some people cannot recognize their own skin tone. According to author Cillian O'Connor colors to avoid for the darker skin Brown is pretty much a no-go. Much like how pastel shades wash out paler skin tones, brown won’t contrast clearly enough with your skin, resulting in your look ‘bleeding’ into your skin. While black and navy are two key colors in any man’s wardrobe (making up the majority of formal and corporate wear), it’s worth keeping these to a minimum to really make the most of your especially versatile skin tone. Fair or pale skin the colors to avoid, given that, depending on lighting, you (or, at least, this author) could be mistaken for a cadaver, you’ll want to avoid an abundance of soft, pastel shades or bright colors as they’ll only wash you out. This goes for neutrals too; rather than white, light beige and stone, choose richer shades such as sand, camel, khaki and slate grey. Red, though not best suited to the ruddier, rosy-cheeked pale person, can work for more alabaster pale skin tones when worn in bold shades like crimson. “I’d recommend basing your outfit around darker colors like navy, brown and black, which you can combine with lighter colors,” Olive or medium skin colors to avoid, since your skin will likely have yellow or green undertones, you should avoid wearing shades of these colors that are too close to your skin tone.

STI College Caloocan

3

Overview of the current state of the technology TheAugmented reality (AR) is a type of interactive, reality-based display environment that takes the capabilities of computer-generated display, sound, text and effects to enhance the user's real-world experience. Augmented reality has many different implementation models and applications, but its primary objective is to provide a rich audiovisual experience. AR works by employing computerized simulation and techniques such as image and speech recognition, animation, head-mounted and hand-held devices and powered display environments to add a virtual display on top of real images and surroundings. The smart mirror, also known as a magic mirror, consists of a transparent mirror and a screen behind the mirror. It shows important information such as clothes. so that the clothes appear to float in the mirror glass. Updates may include new clothes. To display that information, a display should go behind the mirror. It can be a TV, computer screen, or tablet. Framed and frameless styles are both popular, depending on the style of living space. A decision support system (DSS) is an information system that supports business or organization work activities. DSS serve the levels of management, operation and planning of an organization (often in the middle and higher management) and help people make decisions about problems that can be rapidly changing and not easily defined. this is like a smart mirror that the system will help to speed up the selection or give a suggestion to people. Fashion is a popular style, especially in clothing, footwear, lifestyle, accessories, makeup, hairstyle and body. Fashion is a distinctive and often constant trend in the style in which people present themselves. A fashion can become the prevailing style in behavior or manifest the newest creations of designers, technologists, engineers, and design managers. Because the more technical term costume is regularly linked to the term "fashion", the use of the former has been relegated to special senses like fancy-dress or masquerade wear, while the word "fashion" often refers to clothing, including the study of clothing. Although aspects of fashion can be feminine or masculine, STI College Caloocan

4

Objectives of the study General Objectives This study is to develop an Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection. The Kinect device that can be used in selecting fashion outfit that suit on the user’s skin tone using hand gestures. This smart mirror will display fashion outfit selection in Augmented Reality.

Specifically, the study aims the following: 1. To develop a module that will analyze the scanned user's skin tone. The system will scan the dorsal surface of a hand to analyze the user skin tone if it’s white, brown or dark. The fashion designer agreed that matching of fashion outfit is based on the appearance of a person. 2. The system will decide the matched fashion outfit based on the user’s skin tone. In augmented reality (AR) format, the system will show all possible fashion outfits (upper and lower body) that suits for the users which display in a smart mirror.

3. To develop a calibrated fashion outfit based on the users’ positions, poses and gestures using Kinect. The augmented reality outfit will automatically fit and it follows the movement of the users. Using Kinect technology, it captures the body movement then the system will response accordingly.

STI College Caloocan

5

Scope and limitations of the study

Scope 

Captured Skin Tone Module This module will scan the dorsal hand that will check and identify the skin tone using a type of camera to get it the value or the system's interface, So the system can better understand what the user has to do with the user's skin tone so that the system to have provided suggestion to user.



Fashion Outfit Module The system will show the suggested fashion outfit that match to the user’s skin tone color. The augmented reality fashion outfit can be fit and move based on the user’s body using Kinect device.



Captured Fashion Outfit This module is additional features of the system. It will capture the user’s image based on the suggested outfit. The system administrator can view it for future uses.

Limitation Limited wearable Size. It cannot recognize exact body frames or outfit Sizes,the system cannot recognize the size of XS, S, M, L, XL. Event outfit. This program cannot identify if what event or occasion purpose of the Captured outfit is used. The system has limited resources since the researchers are only focused on the fashion outfit itself. It cannot cover cosmetics side and hair designs. The system cannot display any suggestion about enhancement of skin tone. Brand name of the fashion outfit is not part of the system. Single Outfit Selection. The system will not allow the user to select or fit more than one cloth at the same time.

STI College Caloocan

6

Summary This study is to build an, An Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection, the system consists of multiple task including extraction of user’s different body parts from video stream, alignment of models, detecting and resizing input dress images and dress up using Kinect device. We also have used 3D locations for positioning, and system so that user can control to change and choose the clothes. By integrating all the techniques, our system will produce prominently credible and well fitted dress up results. Shortly after the program is started the system is ready to accept users. The users can try on the garment and can interact with the system by utilizing calibrating T-pose, combination with a screen, a so-called virtual mirror. The next part treats the general tracking process, After the program is fully started and the mirrored camera stream is appearing on the screen, it is ready to useby users. The user is prompted to move the body pose. After the pose has been executed and measuring is finished, the registration is complete. The user is now able to interact with the system and choose between the garments. The user has to select the desired garment. This file storing details on the cloth positions for dress up.

STI College Caloocan

7

LITERATURE REVIEW

Review of related literature, studies or systems This chapter presents the related literature and studies after the thorough and indepth search done by the researchers. The theories behind the Augmented Reality System, Smart Mirror Tech, DSS, Fashion outfit system. This will also present the synthesis of the art, theoretical and conceptual framework to fully understand the research to be done and lastly the definition of terms for better comprehension of the study. Foreign Studies Augmented Reality System Augmented Reality-Based Magic Mirror Fashion Coordination System. Augmented reality technology synthesizes images of virtual objects with real world scenes. Existing fashion coordination systems employ a virtual environment and QR codes. This study proposes a coordination system that can effectively coordinate diverse fashions that fit the user’s body, using an augmented reality-based magic mirror. When the coordination system is not activated, an unaltered image of the user is displayed. When the user approaches or touches the display, they will activate the coordination system. The information expressed on the screen is composed of real-time 3D outfits, make-up, and a hair simulation function, which can immediately provide the fashion information requested by the user. Hair design involves the overall practice of planning a new hairstyle. Morphological characteristics such as length, color, and wave are all elements of hair design. Formal characteristics such as order, balance, and harmony are also applied to the principles of hair design. Objectivity properly creates beauty by ordering some design elements through its principles. Hair color can change a person’s image by increasing or decreasing the volume or making up a skin color. It can also establish a hairline for a person’s face. Using bright and warm colors on the parietal area or bangs creates an effect that makes the subject’s face appear longer. When dark and cold colors are used on the side of the head, it narrows the apparent length. However, when bright and warm colors are applied to the same point, it makes the subject’s face appear wider. Hairstyle color changes play a very important role in overall impressions STI College Caloocan

8

by focusing peoples’ eyes. Various colors can be selected according to the preferred image and season, using the general characteristics of the individual colors. Users can employ their own texture or combine it with others, depending on their personality. The magic mirror fashion coordination system features mirror-like display processes that utilize user recognition, face styles, make-up, and dress fitting simulations. An edge detection method was required to facilitate user recognition. Canny Edge, Laplacian, and Log (Laplacian of Gaussian) are representative edge detection methods. The Canny Edge method eliminates image noise using a Gaussian mask, and calculates leaning and strength with x and y axes. A strong point of the Canny Edge method is that it can extract strong edges that are not sensitive to noise. A method of extracting edges through the first order derivative shows sensitive reactions to vertical lines and diagonal lines. To supplement this, the second order derivative searches for closed curve-shaped edge contours. (M. Kim,2014) Digital Clothing and i-Fashion System. The goal of digital clothing technology is to reproduce clothes in 3D using CG. Therefore, digital clothing technology has the potential to advance electronic innovations in fashion. Clothing production methods can be divided into analog, semi-digital, and digital methods. The analog clothing method employs conventional fashion design and garment manufacturing techniques, using panel (pattern) production to produce clothes by cutting and sewing the prepared panels. Semidigital clothing methods integrate digital technology into various analog clothing processes. With this system, fashion design is conducted with design CAD software, and panel production is achieved with pattern CAD software. Digital clothing methods combine a 3D garment reproduction technique with the semi-digital method. With the addition of this new technique, garment design problems can be identified prior to the manufacturing process, allowing corrections to be made to the panels. Thus, digital technology has shortened the fashion item production process, and has been used for the planning of various products shows a digital clothing dress production method, with a computer-based dress reproduction stage added to a semi-digital method. (BioLogic, 2016)

STI College Caloocan

9

Smart Mirror Technology FitMirror: A Smart Mirror for Positive Affect in Everyday User Morning Routines.This study will discuss the concept of a smart mirror for healthier living, the FitMirror. Many people have serious problems to get up after sleeping, to get motivated for the day, or are tired and in a bad mood in the morning. The goal of FitMirror is to positively affect the user’s feelings by increasing his/her motivation, mood and feeling of fitness. While concepts for these isolated problems exist, none of these combine them into one system. FitMirror is implemented to combine them and evaluate them in a study. It consists of a monitor with spy-foil, a Microsoft Kinect v2 and a Wii Balance Board and can recognize users and their gestures with these elements. Several hypotheses about the system regarding motivation, fun, difficulty and getting awake were investigated. Participants were grouped by the factors sportspersons and morning persons to investigate the effect based on these aspects. Results show that FitMirror can help users get awake in the morning, raise their motivation to do sports and motivate them for the day. In recent years, companies1 and scientists have developed smart mirrors, whether in a smart home context, only to display information, or for consumers in boutiques or shopping malls. These mirrors were only developed to either show general information, or 3D visualizations of a person. So far, no smart mirror that offers added value has been developed. Moreover, none of the already mentioned mirrors is able to recognize fitness exercises and offer interaction. This paper presents a smart mirror that provides a combination of several functions and input modalities It includes displaying information, such as weight and other health data, as well as recognition of sports exercises for affecting the users’ feelings positively and motivate them. This raises the question how the feeling of a user can be manipulated. The implemented smart-mirror, the so-called FitMirror, provides the user with normal and fun exercises. These exercises are intended to increase the motivation, happiness and fitness of the user after getting up in the morning. The concept of FitMirror should be regarded as part of a smart home. Before it is described in detail, related work that is relevant for the concept of FitMirror and the underlying technology is presented. These projects are the basis for the concept of FitMirror and show what kind of technologies are available. Afterward, the concept is presented and evaluation results are reported (Besserer, 2016) STI College Caloocan

10

A smart interactive mirror system and present its benefits in decision support for cosmetic product selection. This studyis to contribute to the design of a real mirror system in which the interface is used for virtual application of makeup products. Makeup product selection is challenging. There are different kinds of makeup products and there has been an explosion in the number of makeup brands. Also, there are many different shades and colors for the eyelashes, mascara, nail-polish, lipsticks in a store that makes the makeup selection process overwhelming to the user. Moreover, in order to select a makeup product, the user often needs to be assured that its application will suit her skin tone etc. In this motivation, we present a sensory augmented smart interaction mirror (SIM) system, where the user will be able to apply different makeup products on his/her virtual 3D face in a natural way and see the results. (Rahman, 2010). Decision Support System Analytica. This study makes decision-making transparent and easy with its influence diagrams and powerful built-in probabilistic analysis. Influence diagrams to make it easy not only to understand the model but also to distinguish key decisionmaking variables and their relationships. Built-in tools for managing risk and uncertainty include Monte Carlo and Latin Hypercube. Scenario and sensitivity analysis let decisionmakers know which variables are most important and why. There were two main features that originally attracted me to Analytica: its influence diagram interface for graphically laying out the flow of model logic, and its Intelligent Arrays technology which permits array abstraction. Both of these features integrate with each other to give a modeler the ability to structure and communicate model logic in the most concise and informative way possible. The added benefit of the Intelligent Arrays is that it supports the extension of model dimensionality with little recoding, if any at all. The other indispensable features of Analytica are its support for Monte Carlo simulation, built-in graphing tools, and support for connecting to external data sources. By comparison to a spreadsheet, models of similar complexity require anywhere from 1/4 to 1/2 the amount of time to build in Analytica. Analytica really excels with communicating complex ideas. The graphical layout of the influence diagrams enhanced with control UI elements supports clients more readily comprehending important model details, especially if they were not directly involved with the creation of the model. I frequently take executives through STI College Caloocan

11

very detailed discussions of risk analysis without preparing post processed and simplified information or additional slide decks. (Lumina D.S., 1992) River Logic. This study is a privately held technology firm with corporate offices in Dallas, Texas. River Logic’s Platform enables organizations and enterprises to make more impactful decisions and optimize overall business performance by understanding how to best utilize resources (both physical and intangible resources). Beginning as a leader in prescriptive analytics (optimization), River Logic is now the most competitive advanced analytics platform. It sits at the forefront of technology innovators with its unique ability to offer planning and optimization applications in addition to a platform on which companies can build custom, unique solutions. We believe there is a better way to make decisions today. The power of advanced analytics-based decision making should be in the hands of the people driving the business, not someone in IT or someone sitting behind a desk crunching number. Therefore, our mission is to focus on helping customers and business partners increase their performance, decision-making agility and ability to predict business outcomes through accessible, easy-to-digest prescriptive analytics. Unlike siloed point solutions, our next-generation, cloud-based applications uniquely enable business professionals to collaboratively solve inexplicable or overlooked strategic and tactical business problems by evaluating unlimited what-if scenarios. Business users are empowered to make the most informed, optimal decisions that will drive benefits like additional profit equal to 2-5% of annual revenue, improved service levels and increased shareholder value. River Logic also exposes hidden opportunities while respecting the constraints and variables of a business. We offer the only solution that allows business users to ask unlimited questions regarding scenarios (similar to the way we ask Google questions), receive optimal and feasible plans and collaborate around / track the performance of those plans. This allows companies to have a cross-functional understanding of how to appropriately manage tradeoffs and find opportunities for profit improvement. We encourage a collaborative planning process that crosses all functional silos, thus ensuring that plans no matter the industry or use case a) are tracked against performance metrics and b) are in line w/ overarching business goals and objectives. (River Logic Inc., 2000)

STI College Caloocan

12

Fashion Outfit System Outfit Recommendation System Based on Deep Learning. This study proposes an outfit recommendation system based on deep learning. Our goal is to use the system not only to judge an outfit if it is good or not but also to recommend good outfit to users when it is given a pool of cloth items. Our proposed model includes two parts: one is feature extractor based on ResNet-50, and the other is a binary classifier which is to classify the outfits into good ones and bad ones. Since our model is based on deep learning, it is necessary to use huge data to train the model. We collected a dataset which consists of 409,776 outfits with 644,192 items from the famous fashion website called Polyvore.com. With this dataset, we trained our model and the performance of it is over 84%. And our model can also recommend daily outfit to users. In this work, we presented an outfit recommendation system based on deep learning. We created a novel dataset based on outfits rather than cloth items by selecting suitable outfits from the fashion website named Polyvore.com. We proposed a model that uses ResNet as a feature extractor and a 2-layer MLP as a binary classifier. Since the ResNet is already well trained, we put our attention to the training of our binary classifier. We tried several impossible models and found out that the 4096-fully-connected model performs best. Our model can judge if an outfit is good or bad with high performance. Also, it is can be used in the outfit recommendation system, which is given a pool of cloth items it can recommend the users the best outfit generated from the input items. (Ying Huang, 2017).

STI College Caloocan

13

Mining Fashion Outfit Composition Using an End-to-End Deep Learning Approach on Set Data. In this paper, we propose a machine learning system to compose fashion outfits automatically. The core of the proposed automatic composition system is to score fashion outfit candidates based on the appearances and meta-data. We propose to leverage outfit popularity on fashion-oriented websites to supervise the scoring component. The scoring component is a multi-modal multi-instance deep learning system that evaluates instance aesthetics and set compatibility simultaneously. In order to train and evaluate the proposed composition system, we have collected a large-scale fashion outfit dataset with 195K outfits and 368K fashion items from Polyvore. Although the fashion outfit scoring and composition is rather challenging, we have achieved an AUC of 85% for the scoring component, and an accuracy of 77% for a constrained composition task. (R.Kumar, 2017). Synthesis "MirrorTech”: A Smart Interactive Mirror with Decision Support in fashion outfit selection" is relevant to the Smart Mirror because it the same uses displaying time and date and it has also the concept about on using speech for audio feedback and speech command to interact with. The proposed software differs from foreign studies "Augmented Rendering of make-up features in a Smart Mirror System for Decision Support in Cosmetic Products Selection" because it benefits in displaying an augmented 3D representation of the user with make-up features. In this approach the user is able to view the possible outcomes of different makeup applications in the smart mirror without affecting the real appearance in the process. The use of technology is increasing day by day, we all depend on technology, and we use various technologies to accomplish specific tasks in our lives. Today we have various emerging technologies that impact our lives in different ways. Technology is being implemented in almost every aspect of our lives. We all know that technology helps us to do things far greater than our ability and resources.

STI College Caloocan

14

“MIRROR TECH” AN AUGMENTED REALITY SYSTEM USING SMART MIRROR TECHNOLOGY WITH DECISION SUPPORT IN FASHION OUTFIT SELECTION.

Overview of the project Mirror Tech is a device that aims to guide and to assist what users need in outfit. It could also help them to automatically measure what clothes that fits to the user’s body, using Kinect device synthesis for hand gesture and calibrating pose to interact with the device. Mirror Tech can also upload new clothes by the admin. The admin will upload new clothes to update the device. This kind of device, Mirror Tech, can also help the user to use the device better.

System design specification Mirror Tech use prototype model. Applied with different phases. Our main focus in developing our tech is the system and the hardware wherein we will first build, test it then rework and add the necessary touches until our desired prototype is finally achieved. The Prototyping Model is a Systems Development Methodology (SDM) within which a paradigm output (or an early approximation of a final system or product) is constructed, tested, and then reworked. It is done till an appropriate paradigm is achieved to help develop the entire system or product. This model works best in situations when all the details or requirements are not known well in advance. It is majorly a trial-and-error process which works in an iterative fashion.

Requirements Gathering

Quick Design

Start

Building Prototype

End

Engineer Product

Refining Prototype

Customer Evaluation

Figure 1. Prototype Model (MirrorTech) STI College Caloocan

15

In Requirements Gathering and Analysis Phase, the mirror tech uses different Hardware device and Software Specification. the overall system has two main components, a hardware component and a software component. The hardware model is a one-way mirror covering an LCD screen, both of which approximately measure 40” diagonally. All components are encapsulated within a wooden frame. On the outside, the hardware is encapsulated within a wooden frame. On the front, a one-way mirror and web camera are placed in front of an LCD monitor. This way, the system can act as a mirror when not currently in use, while the LCD projects through the mirror when in use. It also the wooden frame has a bezel on the front which the mirror and LCD panel are pressed against. Cutouts for dowels are added in line with the back of the LCD panel to keep the components snug against the bezel. We also have a microphone and speaker attached for audio input and output. In order to use the voice control features of your smart-mirror, you will also need a USB microphone. The goal of our project is to create an open platform for development, and all software components must fit to that goal. In Software Specification, the mirror assembled, the mirror will display the data that support the dress up system.

In Quick Designwhenrequirements are known, a preliminary design or quick design for the system is created. In this stage, the proponents started to create interfaces of a device and planning of how the device will interact to the user. In building prototype this phase new invention ideas become workable product samples as a result of invention lands prototyping process. Following the design and engineering process, which lay out product ideas digitally, the prototyping process create, a tangible sample that can be interacted with and tested. Throughout the prototyping process, invention land builders work closely with designers, but they have the flexibility to make changes if a new product idea is not fitting together or functioning properly. For this reason, interactive testing plays a major role in building a working prototype.

STI College Caloocan

16

Figure 2(MirrorTech) Homepage

In User evaluationthe proposed system is presented to the user for thorough evaluation of the prototype to recognize its strengths and weaknesses such as what it to be added orremoved. Next step, the proposed device will present to the chosen client (Resources for the F&S Fashion) for thorough evaluation of the prototype to recognize its strengths and weaknesses and in order to allow the client to suggests more features to the proposed device in order for it to work better and will benefit for them.

In Refining prototypeonce the user evaluates the prototype and if he is not satisfied, the current prototype is refined according to the requirements. That is, a new prototype is developed with the additional information provided by the user. The new prototype is evaluated just like the previous prototype. This process continues until all the requirements specified by the user are met. Once the user is satisfied with the developed prototype, a final system is developed on the basis of the final prototype. Once the client will not be satisfied to our device, the current prototype is refined according to the requirements.

STI College Caloocan

17

In Engineer product the final system is evaluated thoroughly followed by the routine maintenance on regular basis for preventing large-scale failures and minimizing downtime. After all the processes that the proposed device which is the Mirror tech is finished and all the requirements by the client are met, the final prototype will accept by them.

Hardware/Software

Hardware

Figure 3 The scenario of using dynamic fitting room Sony 40 Inch Television

Television (TV) is a telecommunication medium used for transmitting moving images in monochrome (black and white), or in color, and in two or three dimensions and sound.

Figure 4 Body index and gesture. STI College Caloocan

18

Kinect V2

Kinect is Microsoft’s motion sensor add-on for the Xbox 360 gaming console. The device provides a natural user interface (NUI) that allows users to interact intuitively and without any intermediary device, such as a controller.

Figure 5 Scanner skintone Logitech HD Pro Webcam C920e, 1080p camera The most popular use of webcams is the establishment of video links, permitting computers to act as videophones or videoconference stations. Other popular uses include security surveillance, computer vision, video broadcasting, and for recording social videos.

Figure 6 Main function. STI College Caloocan

19

Computer A computer, sometimes called a notebook computer by manufacturers, is a battery- or AC-powered personal computer generally smaller than a briefcase that can easily be transported and conveniently used in temporary spaces such as on airplanes, in libraries, temporary offices, and at meetings. A laptop typically weighs less than 5 pounds and is 3 inches or less in thickness.

1. 4gb Ram. 2. OS windows 10.. 3. 500gb Hard Drive. 4. Dual core 1.7 – 2.5 GHz.

Software

Visual Studio 2013 C# (version 6.0)

C# is a multi-paradigm programming language encompassing strong typing, imperative, declarative, functional, generic, object-oriented (class-based), and component-oriented programming disciplines. It was developed by Microsoft within its .NET initiative and later approved as a standard by Ecma (ECMA-334) and ISO (ISO/IEC 23270:2006). C# is one of the programming languages designed for the Common Language Infrastructure.

Unity version (2017 .4.1f1) Personal Unity Personal is a great place for beginners and hobbyists to get started. It includes access to all core game engine features, continuous updates, beta releases, and all publishing platforms.

Python v3.7.2 Python is a high-level programming language designed to be easy to read and simple to implement. It is open source, which means it is free to use, even for commercial applications. Python can run on Mac, Windows, and Unix systems and has also been ported to Java and .NET virtual machines.

STI College Caloocan

20

RESULT AND DISCUSSIONS The proponents conducted a survey on our client’s customer; the survey was primarily based on their experiences on using smart mirror and if the user think that using technology will help them fitting clothes process quicker and simpler.

STI COLLEGE CALOOCAN “MIRROR TECH” AN AUGMENTED REALITY SYSTEM USING SMARTMIRROR TECHNOLOGY WITH DECISION SUPPORT IN FASHION OUTFIT.

Please check the appropriate box to each corresponding detail.

Strongly Agree (1)

Agree (2)

Neutral (3)

Strongly Disagree (4)

Details

1

Disagree (5)

2

3

4

5

The Mirror Tech Skin Tone Scanner 1. Displays captured dorsal surface of my hand 2. Provides efficient skin color result that matches my skin tone 3. Can display the text of precise skin tone. The Mirror Tech Skin Tone Analyzer 4. Identifies my skin tone if it’s white, brown, or black. 5. Displays fashion outfits that suits for me 6. Offersapproximate designs of outfits based on skin tone. The Mirror Tech Kinect Device 7. Its user friendly and fun to use 8. Helps me to control mirror tech quicker and simpler 9. Natural and convenient to use. The Mirror Tech Decision Support 10. Can suggest fashion outfit in augmented reality.

Name (Optional) STI College Caloocan

Date 21

The questionnaire above was used in the survey. The proponents select 50 random F&S tailor’s respondents. The proponents want to identify if they the users have any idea on how this kind of technology works. Details

Strongly Agree

Agree

1 2 3 4 5 6 7 8 9 10

50 45 50 48 38 35 48 45 47 45

0 5 0 2 12 15 2 5 3 5

Table #1 Survey Result The first 3 questions will be based on the first objective. The first objective is to develop a skin tone scanner able to knows what is him/her skin tone. In the table 1 results that 1-3 have a result of 50, 45, 50 that is strongly agree that the mirror tech can scan the users skin tone. so, all of the correspondent find that the mirror tech working precisely.

The second 3 questions will be based on the second objective. The second objective is to analyze the skin tone to display the suggested fashion outfit based on the user’s skin tone. In the table 1 results that 4-6 have a result of 48, 38, 35 that is strongly agree that the mirror tech can analyze the users skin tone to display fashion outfit. The correspondent test that it is working.

The third 4 questions will be based on the third objective. The first objective is using a Kinect device. In the table 1 results that 7-10 have a result of 48, 45, 47, 45 that is strongly agree that the mirror tech is using a Kinect device. All of the correspondent saw that the mirror tech is smooth to use.

STI College Caloocan

22

CONCLUSION AND RECOMMENDATIONS In this work we introduce an augmented reality dressing system which only requires having a front image for each product to superimpose it onto the user and the 3D graphics of the product seem to be relatively satisfactory and practical for many uses. We presented the methodology that we use to align the models with the user and we tested our procedure under different conditions. The experiments have resulted with acceptable performance rates for regular postures After an introduction, the related work was presented real time tracking technologies up to an overview of comparable augmented reality dress-up system. Subsequently a closer look on the technologies and frameworks that were used for the implementation of the Augmented Reality Dressing System was taken. After this the different aspects of the design process up to the construction of the garment models were highlighted. This is followed by the implementation, describing the cloth colliders and the behavior of the garment, for instance. In the last section the tests were executed, also discussing the output, the appearance and the interaction with the augmented reality dressing system. Overall, the presented Augmented Reality System using Smart Mirror Technology with Decision Support in Fashion Outfit Selection seems to be a good solution. The approach and results presented here will benefit designers and developers of similar applications in the future. Responsive Mirror is an instance of an intelligent user interface which users implicitly control, allowing the technology toseamlessly support user goals without disruption.

STI College Caloocan

23

MONTH ACTIVITY

JUNE

1 2 3 4 1

JULY

AUG.

SEPT

2 3 4 1 2 3 4 1 2 3 4 1

OCT.

NOV.

DEC.

JAN.

FEB.

2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

Progress of the System/Software Concept of Requirements and Operations Initial Estimation Data Gathering and Scenario Developing the system Integration Testing and Debugging Documents Writing and Submission of Chapter 1,2,3 Oral Defense Revision of Documents

FINISHED ON PROGRESS TARGET

STI College Caloocan

24

Quantity

Specifics

1 1 5 1 2 1 8

Sony 42” inch TV Logitech HD Pro Webcam C920 Plywood PW ½ Kinect V2 LED Strip (5meters) Camera A4tech Spray paint (pylox)

Approximate Cost 15,000 5,000 1,500 3,000 900 500 1000

TOTAL

Actual Cost 30,000 4,025 1,650 3,000 1100 450 1360

41,585

Prepared by:

Jinky B. Fuentes

Renz Adriane E. Ocampo

Ray Clark D. Sulit

Joan G. Villanueva

Noted by: Sisa B. Fuentes

Myrna Bustamante

Josephine C. De Jesus

Shirley G. Villanueva

Approved by:

Mr. Bernard A. Mendiola

STI College Caloocan

Mrs. Marilou F. Bartolome

25

Curriculum Vitae of

RENZ ADRIANE E. OCAMPO #10 C3 Road Navotas City [email protected] 09104257894

Level Tertiary High School Elementary

EDUCATIONAL BACKGROUND Inclusive Dates Name of school/ Institution 2015-2019 STI College Caloocan 2010-2014 Navotas National High School 2010 Baptist Elementary School

PROFESSIONAL OR VOLUNTEER EXPERIENCE Nature of Experience/ Name and Address of Company or Inclusive Dates Job Title Organization

Inclusive Dates

SKILLS

AFFILIATIONS Name of Organization

Position

SKILLS Level of Competency

Date Acquired

TRAININGS, SEMINARS OR WORKSHOP ATTENDED Inclusive Dates Title of Training, Seminar or Workshop

STI College <School Name>

26

Curriculum Vitae of

RAY CLARK D. SULIT #181 Marcos St. IslaValenzuela City [email protected] 09104616897

Level Tertiary High School Elementary

EDUCATIONAL BACKGROUND Inclusive Dates Name of school/ Institution 2015-2019 STI College Caloocan 2010-2014 Polo National High School 2010

Isla Elementary School

PROFESSIONAL OR VOLUNTEER EXPERIENCE Nature of Experience/ Name and Address of Company or Inclusive Dates Job Title Organization

Inclusive Dates

SKILLS

AFFILIATIONS Name of Organization

Position

SKILLS Level of Competency

Date Acquired

TRAININGS, SEMINARS OR WORKSHOP ATTENDED Inclusive Dates Title of Training, Seminar or Workshop

STI College <School Name>

27

Curriculum Vitae of

JINKY B. FUENTES #8 Flerida St. Acacia Malabon City [email protected] 09463996404

Level Tertiary High School Elementary

EDUCATIONAL BACKGROUND Inclusive Dates Name of school/ Institution 2015-2019 STI College Caloocan 2010-2014 Tinajeros national High School (Acacia Annex) 2010 Acacia elementary School

PROFESSIONAL OR VOLUNTEER EXPERIENCE Nature of Experience/ Name and Address of Company or Inclusive Dates Job Title Organization

Inclusive Dates

SKILLS

AFFILIATIONS Name of Organization

Position

SKILLS Level of Competency

Date Acquired

TRAININGS, SEMINARS OR WORKSHOP ATTENDED Inclusive Dates Title of Training, Seminar or Workshop

STI College <School Name>

28

Curriculum Vitae of

JOAN G. VILLANUEVA 631 J.FelipeSt.Sapang Caloocan City [email protected] 09569759121/2851156

EDUCATIONAL BACKGROUND Inclusive Dates Name of school/ Institution 2015-2019 STI College Caloocan 2010-2014 Caloocan High School

Level Tertiary High School Elementary

2010

AndressBonifacioElem.School

PROFESSIONAL OR VOLUNTEER EXPERIENCE Nature of Experience/ Name and Address of Company or Inclusive Dates Job Title Organization

Inclusive Dates

SKILLS

AFFILIATIONS Name of Organization

Position

SKILLS Level of Competency

Date Acquired

TRAININGS, SEMINARS OR WORKSHOP ATTENDED

STI College <School Name>

29

REFERRENCES

https://pdfs.semanticscholar.org/a59f/1a5c7033bd2e94f78f7d558077874295e8ff.pdf https://pdfs.semanticscholar.org/a59f/1a5c7033bd2e94f78f7d558077874295e8ff.pdf https://www.researchgate.net/.../284438142_Augmented_Reality_Fashion_Apparel_Sim https://www.researchgate.net/.../273306624_Magic_Mirror_Fashion_Coordination_Syst https://www.researchgate.net/.../261308641_Mirracle_An_augmented_reality_magic https://www.researchgate.net/.../309409131_FitMirror_A_Smart_Mirror_For_Positive_... https://www.uni-ulm.de/fileadmin/website_uni_ulm/.../Besserer2016_preprint.pdf

dl.acm.org/ft_gateway.cfm?id=3011265&ftid=1813205&dwn=1&CFID. https://www.semanticscholar.org/.../Fitmirror%3A...smart-mirror-for-positive-affect. https://www.ijirset.com/upload/2018/ncetdst/45_Smart%20Mirror%20final.pd https://apps.dtic.mil/dtic/tr/fulltext/u2/a166327.pdf

https://www.nap.edu/read/13040/chapter/5

https://www.nap.edu/read/13062/chapter/15 www.lumina.com/technology/influence-diagrams https://www.lumina.com/.../decision-tree-software-using-analytica-as-a-powerful-alter.. https://download.atlantis-press.com/article/25880175.pdf

https://www.atlantis-press.com/proceedings/iccia-17/25880175

https://www.researchgate.net/.../325973475_Fashion_Outfit_Composition_by_Deep_Le..

STI College <School Name>

30

https://www.researchgate.net/.../318798409_Outfit_Recommendation_System_Based_on . https://www.groundai.com/.../fashionnet-personalized-outfit-recommendation-with-de...

htts://www.emeraldinsight.com/doi/full/10.1108/IJCST-09-2015-0100?fullSc=1 https://www.facelet.com/.../smart-fitting-rooms-how-they-work-and-why-stores-need-t...

https://www.cbinsights.com/research/fashion-tech-futuretrends/https://www.techemergence.com/virtual-mirrors-and-computer-vision

STI College <School Name>

31