Skip to content

imedvedev75/VisualProductRecommendations

Repository files navigation

Visual Product Recommendations

The MVP of Visual Product Recommendations is designed to be a game changer for the shopping experience. By leveraging advanced AI technologies, the app provides product recommendations based on user-provided images, focusing primarily on clothes and shoes.

Overview

The project employs a fine-tuned ImageNet model, specifically adapted to recognize a wide range of clothing and footwear. The dataset, collected from the Amazon website using a custom crawler, is enriched through computer vision algorithms to enhance prediction quality. The AI model operates as a REST service.

Key Features

  • Image-Based Recommendations: Users can take photos of clothes or shoes, and the app intelligently recommends similar products.
  • AI-Driven Insights: Utilizes a fine-tuned ImageNet model, specifically optimized for high accuracy in clothing and footwear categorization.
  • Data Augmentation: The dataset is augmented using computer vision techniques, to increase the quality of predictions.
  • RESTful AI Model: The AI model is served as a REST service, allowing for easy integration and scalable architecture.
  • User-Friendly Interface: The Android app provides a straightforward and intuitive interface for capturing photos and viewing recommendations.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published