Build apps with state-of-the-art computer vision
summary
Project Florence powers Azure Cognitive Services for Vision to generate image captions and dense captions, extract shared image and text embeddings for vector search, and run segmentation and video summarization/frame-locator pipelines that successfully find frames (examples shown with factory and retail footage). Integration with Axis smart cameras sends motion-triggered clips to Azure Blob, indexing via Azure Cognitive Search private preview. Model customization supports few-shot learning (as little as one image per label) for fine-grain produce and shelf-planogram matching; Reddit and Blue Yonder use these APIs in production or pilots.
AI-written summary · about
chapters (12)
- 0:00Introduction to Session
- 0:51Project Florence Overview
- 5:11Digital Asset Management
- 6:11Image Search & Analysis
- 11:03Image Segmentation
- 11:52API SDK Demonstration
- 15:11Safety and Security
- 19:05Smart Camera Integration
- 26:38Modern Retail Automation
- 28:34Shelf Analysis and Compliance
- 30:51Retail Demo and Success
- 37:53Q&A and Conclusion
related talks
57mEasier AI allows everyone to build smart apps
01h 33mComputer Vision in Azure
46mBuilding accessible experiences with AI
01h 03mEasier AI allows everyone to build smart apps
01h 05mMicrosoft Cognitive Services: Making AI Easy
01h 00m5 Levels of AI in Microsoft Azure
29mApplied Custom Vision
41mAI + Machine Learning for .NET Developers