computer vision

Exploring Real-World Computer Vision: Deployments, Edge Models, and Current Challenges

A developer who built the food recognition model for MyFitnessPal is now asking a grounded question: what are people actually deploying in industry today?

3 min readMachine Learning
From Machine Learning

I've built a lot of ML systems over the years, mainly computer vision models optimised to run on mobile phones. For example, my previous company built the food recognition model for MyFitnessPal.

I'm interested in what people are actually deploying in industry now. Are edge models still a big part of your work, are you hosting your own models, or are you mostly sending requests to APIs? What's driving that choice?

Read the original at Machine Learning