Augmented Reality Based Fashion Simulation

Project Title

Augmented Reality Based Fashion Simulation

Project Area of Specialization Augmented and Virtual RealityProject Summary

Companies using Augmented Reality benefit from increased revenue when customers experience products before purchasing them, it deepens their sense of ownership. Customers who feel as though they already own the product are likely to purchase. The AR experience is also a great opportunity for companies to educate customers and offer more in-depth content beyond the packaging and label. It also presents the perfect opportunity to upsell companion products or accessories.

Augmented Reality based Fashion Simulation project aims to develop the system in which user can select and virtually try on the apparels available in the retail shop, filtered according to his/her body parameters. In this proposed work our main goal is to implement a practical solution that can provide greater satisfaction. This can be achieved by implementing an augmented reality experience in the most realistic way including physics motion to the dress according to the movements made by customer, using depth sense camera.  The proposed system is developed with the focus on :

• 3D models of local cloth designs, to enhance scalability. For system implementation we used a Kinect v2 depth sensor equipped with several sensor that can detect depth using an IR sensor, capture RGB images using a camera sensor.
• The necessary parameters for clothing such as height, shoulder length, arm length and neck-hip length are calculated incorporating with Kinect skeleton joints.
• The developed system consists of an application “Unity 3D” to overlay the 3D cloth model on the user in front of device.
  • Here the cloth model has used 3D designed software “3DMAX” for rigging and creating 3D cloth models. Once it was rigged, the 3D models was imported to the Unity3D project. Then, using configuration menu for humanoid model, it was matched with the joints of human model. The resultant model can be converted into controllable 3D humanoid model.
• The selected model has been overlaid on the top of the user on the video stream in real time. Results will indicate that a depth sensor can be used successfully for the purposed AR virtual dressing room.

Project Objectives

The project helps in achieving following Objectives :

• To display clothes, accessories and 3D models, desktop application is being designed.
• To detect human body parameters, such as waist, shoulders, and other joints, Kinect v2 sensor is connected.
• To capture user body parameters using depth sensor, C# based application is being implemented.
• To capture environment data through Kinect RGB camera and IR depth sensor.
• To create an animated or controllable model, the selected model needs to be rigged.
• To simulate the interaction between the 3D avatars and digital garments.
• To virtual try on apparels available in the application, according to his/her body parameters.

Project Implementation Method

Augmented Based Fashion Simulation is real time application which facilitate user to virtual try different accessories.

• Desktop Application :

3D fitting room is being developed containing the accessories for both male and female i.e. clothes, handbags, watches and caps. Desktop Application is being developed to categorize the gender and accessories according to gender.

• Human Body Detection :
• Kinect v2 based depth sense camera is connected to capture human body parameters such as waist, chest, hip, thigh, knee joints through its RGB camera and IR depth sensor to utilize data
• Dress-up mannequins :

Garments and accessories are captured on real time mannequins, then rigged 3D models will be implemented in unity 3D and customers can virtually try those clothes and accessories.

 This helps in animating the cloth model according to the skeletal movement. Several methods have been used in the literature to achieve this, such as automated rigging methods. Instead of using automated methods, this project uses manual process to verify and build skeletal on the top of cloth model. This is mainly due to the requirement of a humanoid avatar, which matches to skeleton. Here, this project uses cloth model instead of avatar model. We can create a rigged system with 3D Max prebacked. This needs to be a manual work such as aligning the cloth model and skeleton in a way that it gives the effect of real human body.

                Methodology :

 

Design/Implementation :

  

Benefits of the Project

• Reduce Returns Rates
• Improving Personalization
• No contact try-on
• Time-Saving
• No need to go into physical fitting room
• Meet growing shopper expectations
• Customers do not need to afraid of hidden cameras and two-way mirrors
• Risk free zero-liability Trails
• Increase Buyer’s confidence
• Unlimited product portfolio
• Increase buyer’s confidence
• Boosting desires              
• Online users shouldn’t worry about getting something that will not suit on them.            

It allows designers to unleash their creativity in a real-life visualization of designers that could previously only be imagined through 2D sketches.

Technical Details of Final Deliverable

The final product will be :

• Desktop Application containing facilities for both male and female.
• Kinect v2 based depth sense camera to detect the human body
• User would have to stand in front of camera and maintain their positions for virtual trial of clothes and accessories.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther IndustriesCore Technology Augmented & Virtual RealityOther TechnologiesSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	60500
Kinect v2 sensor	Equipment	2	20000	40000
Adapter	Equipment	2	1000	2000
12 GB RAM	Equipment	1	9000	9000
Printing	Miscellaneous	500	5	2500
Thesis Binding	Miscellaneous	2	2000	4000
Thesis boards and posters	Miscellaneous	1	1500	1500
Visits	Miscellaneous	3	500	1500