Skip to content

A personalised Clothing Recommender System which takes a image as an input of Cloth and Recommend Cloth which matches the input

Notifications You must be signed in to change notification settings

Utkarsh-2002/Attire-Advisor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Attire Advisor


Overview

Attire Advisor is a cutting-edge clothing recommendation system that leverages Convolutional Neural Networks (CNN) to analyze user-uploaded images and suggest outfits that match the detected clothing items. Initially focused on image-based input, Attire Advisor recommends similar styles and complementary pieces. Future iterations of the project will expand to include additional inputs, such as occasion and personal style preferences, allowing for more tailored recommendations.


Project Goals

  • Image-Based Recommendations: Use CNNs to analyze images and suggest outfits that align with the user's current style.
  • Enhanced Personalization: Future versions will incorporate inputs like occasion, weather, and personal style to refine outfit suggestions.
  • User-Friendly Interface: Develop an intuitive platform where users can easily upload images and receive outfit recommendations.

Current Features

  • Image Upload: Users can upload an image of their outfit or a specific clothing item.
  • CNN-Based Analysis: The system uses a Convolutional Neural Network to analyze the uploaded image and identify clothing items.
  • Style Matching: Based on the analysis, the system suggests outfits that match the detected style and clothing type.
  • Outfit Gallery: A collection of suggested outfits that users can browse, based on their image input.

Future Features

  • Occasion-Based Filters: Allow users to specify the occasion (e.g., casual, business, formal) to refine recommendations.
  • Style Customization: Include inputs for personal style preferences, such as color choices, preferred brands, and clothing types.
  • Weather Integration: Integrate real-time weather data to suggest weather-appropriate outfits.
  • Shopping Links: Provide direct links to online stores for purchasing recommended items.

Dataset

  • Fashion Dataset: The model is trained on a large dataset of fashion images, categorized by type, style, and occasion.
  • User Images: Collected through user uploads to generate personalized outfit recommendations.

Model Architecture

  • Convolutional Neural Network (CNN): The core of the Attire Advisor's recommendation system, designed to extract features from fashion images.
  • Feature Extraction: The CNN model extracts features from the uploaded image, such as color, pattern, and clothing type.
  • Style Matching Algorithm: Compares extracted features with the fashion dataset to find the closest style matches.

Snapshots

Below are snapshots of some of the features included in the Attire Advisor:

Screenshot (118) Screenshot (175)


Future Work

  • Enhanced AI Recommendations: Implementing advanced machine learning techniques to improve the accuracy of outfit suggestions over time.
  • User Feedback Integration: Allowing users to rate and provide feedback on recommendations to refine future suggestions.
  • Mobile App Development: Expanding the platform to include a dedicated mobile app for on-the-go outfit recommendations.
  • Occasion and Style Inputs: Integrating additional inputs like occasion, style, and weather to provide more personalized recommendations.

About

A personalised Clothing Recommender System which takes a image as an input of Cloth and Recommend Cloth which matches the input

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published