Get personalized fashion recommendations for any occasion!

Inspiration

Our inspiration is the desire to simplify and enhance the process of selecting the perfect outfit for any occasion. We found that many people are confused about finding the right clothing for particular occasions or events.

What it does

Our platform provides personalized fashion recommendations for any occasion, whether it is a casual gathering, a formal event, or anything. Users give input event details, their style preferences, age, gender, and body type, and our algorithm generates suggestions for clothing, accessories, and shoes.

How I built it

I built our platform using a Large language model(LLM) provided by aws partyrock.aws. This Service can create applications using only a prompt and a little bit of editing drag and resize. And this is our prompt

Provide fashion advice tailored to specific occasions or events. Recommend suitable attire, including clothing, accessories, and shoes, considering the theme, venue, and formality level. Ensure suggestions align with current trends and suit the wearer's style preferences and body type.

Additional Edits :
in Outfit Recommendation Section added custom prompt :
"Please recommend a complete outfit for @Occasion , comprising a top, bottom, shoes, and accessories, tailored to @Style Preferences, suitable for @Body Type, @Gender, standing at @Height, and aged @age , all within a budget of @Budget . Ensure to provide clickable links for each item to shop the look on Amazon.in."

in Outfit Visualization Section :
Please create a visual representation of a recommended outfit suitable for @Occasion incorporating @Style Preferences. The outfit should complement a @Body Type body, be appropriate for someone aged @age , with a @Skin tone skin tone and match the @Gender of the individual. The total budget for the outfit should not exceed @Budget. The visual should showcase both the front and back of the full outfit.

in Helping Bot Chat section :
initial message : Ask me about anything about fashion outfit.
prompt : "Pretend you are a fashion designer. The user will now have a conversation with you about @Occasion outfit. If ask irrelevant questions like out of context from fashion tell them "I could not help you with that. Ask me about fashion outfit only."

Challenges I ran into

I didn't run into very difficult challenges in my case is main challenging part was coming up with a good idea and a better prompt , because all complex things are handled by the partyrock.aws in the backend

Accomplishments that I'm proud of

I am proud to have developed a user-friendly platform that empowers individuals to make confident fashion choices for any occasion.

What I learned

I learned how AI solves complex problems in a very simple way

What's next for Get personalized fashion recommendations for any occasion!

In the future, I plan to further enhance our platform by user feedback mechanisms and refining our recommendations with different language models.

Note:

Generated amazon links are not redirecting to the product because it is not giving Realtime links I don't know exactly why. In my opinion, This Generative AI based on historical data or pre-trained LLMs only, can be solved using Retrieval-Augmented Generation (RAG)

RAG: RAG is a powerful technique that combines the strengths of LLMs with the precision of a retrieval system. Like search engine Ai (Copilot) An RAG model augments the traditional generation process by incorporating an external retrieval or search mechanism.

Built With

• cloud-services
• generative-ai