Web AI in industry: How TensorFlow.js has driven what you see on the supermarket shelves
summary
A Brazilian CPG deploys an offline real‑time product detector to replace manual shelf counts, training YOLOv7 on SKU110K plus 5,000 hand‑labeled images for 40 SKUs, converting to TensorFlow.js as a 25 MB model that runs on phones. Colab + Label Studio + a Visual Blocks server let regional teams train and test models; a central team handles governance. Rollout to 20,000 merchandisers yields a 40% productivity gain, single‑digit market‑share lift and three‑digit‑million revenue increase.
AI-written summary · about
chapters (7)
- 0:00Introduction to Web AI
- 1:12Supermarket operational flow
- 3:39Developing the AI solution
- 5:23Model performance and results
- 6:49Empowering regional managers
- 8:35Streamlined AI workflow
- 10:23Impact and conclusion
related talks
41mAI in Web: TensorFlow js
22mVue + TensorFlow.js: Building Your First ML Powered App
39mMachine Learning Magic for Your JavaScript Application
41mTensorFlow.js 101: ML in the Browser and Beyond
29mSuperpowers for next gen web apps: Machine learning
20mTensorFlow.js Bringing Machine Learning to the Web and Beyond
15mMachine learning for next gen web apps with TensorFlow.js
28mMachine Learning on the web with TensorFlow.js