Inspiration-The idea sparked from watching people struggle to find the right makeup shades or test out new cosmetic styles without physically applying and wiping off products repeatedly. Inspired by expert beauty mentors and personalized retail experiences, we wanted to bridge the gap between technology and self-expression, making custom beauty advice instantly accessible to everyone.
What it does-Our AI Assistant Makeup application acts as a personal virtual beauty advisor. It analyzes a user's facial structure, skin tone, and undertones through a live camera feed or photo. It then provides personalized makeup recommendations (foundation, lipstick, eyeshadow, and contour) and lets users virtually "try on" different looks in real-time before making any purchases.
How we built it-We built the application using a modern tech stack centered around artificial intelligence and computer vision. We utilized computer vision models (such as OpenCV or MediaPipe) for real-time facial landmark detection, machine learning algorithms for skin tone classification, and integrated a conversational AI frontend to answer user beauty queries and guide their product choices.
Challenges we ran into-One of our biggest hurdles was managing AI and API credits/rate limits while processing real-time video frames for facial tracking. Additionally, ensuring accurate color rendering across different lighting conditions and diverse skin tones required careful fine-tuning of our machine learning models and image-processing filters.
Accomplishments that we're proud of-We are incredibly proud to have successfully built a real-time virtual try-on engine that maps cosmetics accurately onto user faces without lag. We are also thankful for our mentor's guidance, which helped us streamline our architecture and keep the user experience seamless and intuitive.
What we learned-We learned how to solve complex real-world computer vision challenges, particularly regarding facial recognition, latency optimization for live video, and how AI can democratize the beauty industry by offering personalized advice tailored to individual features.
What's next for AI agri assistant-
1.Expanding our product catalog to feature more diverse cosmetics brands and shades.
2.Introducing an AI-powered outfit and hair color coordination feature to match full-body aesthetics.
3.Optimizing the app for mobile deployment (iOS and Android) to make everyday beauty shopping effortless
Log in or sign up for Devpost to join the conversation.