Virtualrobe

A Fashion AI app that styles and suggest outfits to wear with clothes in your actual wardrobe. Sign up now to join the list of beta tester: SIGN UP NOW

Features

• AI fashion stylist: Automatically suggest outfits to wear from items in your actual wardrobe
• Newsfeed : View the latest trending outfits created on the platform
• Virtual wardrobe: Allows users to upload 2d images of their clothes, shoes and fashion accesories
• Wardrobe/Closet mangement system: Helps to organize your warobe items and also come with a virtual laundry basket
• In-app messaging and search engine
• Push notification: For auto-reminder

Additional features

Note: Not yet added

• Virtual Shopping mall: This serves has in app shopping service/E-commerce
• 3D image uploading engine: To upload wardrobe items in 3D
• Payment system: Planning to integrate with stripe
• Shipping and delivery service: Planning to integrate with Shippo

Development Stack

• Operating system: Android
• Development environment: Android studio
• Machine learning software: Tensorflow
• Midware and Api : Firebase cloud functions
  • Language used for functions: Javascript and Typescript
• Storage: Inititally used amazon S3, currently migrated to Firebase storage
• Database: Firebase Firestore

Read more & References

My white paper on Virtualrobe and also white papers on using Generative Adversarial Networks for fashion

• Virtualrobe White Paper by Seun kola: http://www.ijaist.com/wp-content/plugins/download-attachments/includes/download.php?id=3772
• https://www.technologyreview.com/s/609469/this-ai-learns-your-fashion-sense-and-invents-your-next-outfit/
• https://www.cs.toronto.edu/~urtasun/publications/zhu_etal_iccv17.pdf
• http://on-demand.gputechconf.com/gtc/2018/presentation/s8776-gan-fashion-photo-shoot-garment-to-model-images-using-conditional-gans.pdf