"Turnkey computer vision to extract data from images"
Automates the entire process for building computer vision, allowing developers to get custom vision recognition models from just a model spec.
Founded by Alex Liao
TLDR: Dragoneye helps developers use computer vision without needing to bring any of their own training data or do any CV/ML work themselves.
Check out the demo ⤵️
❓The Problem
Getting data from images today is hard. LMMs allow you to get started quickly but have capability limitations and are slow. On the other hand, building your own vision recognition models requires a ton of time and a ton of expertise.
Alex learned this firsthand when he tried to build his own fashion outfit**-**matching app a few years ago. What started as coding a fun consumer app was quickly taken over by the challenges of building custom image models - annotating data, model evaluation, training, deployment, etc. Before long, he realized that he was basically building a computer vision startup - literally!
✨The Solution
Dragoneye enables developers to build and use cutting-edge CV models simply by specifying the model spec.
Specification → Working Model Automatically: After specifying the model, Dragoneye’s systems scurry away to find the best implementation strategy and then builds it for you!
Continuous Improvement: In the background, Dragoneye constantly keeps the models and tech at the cutting edge so you keep getting the best performance without any work on your end at all.
Grows with your needs: As your needs grow, they grow with you. Simply update your model specification, and Dragoneye will figure out how to make it work best.
