Florence v1.0 is a computer vision foundation model from Microsoft Research that successfully scales different vision and vision-language tasks. Florence v1.0 is now commercially available for customers through Azure AI, Seeing AI (an app for visually challenged people) and other Microsoft services.
Project Florence’s mission is to take the advancements being made in areas such as feature representation learning, transfer learning, and model architecture search and turn them into applications that can empower our partners and customers to achieve more with Azure Cognitive Services.
Microsoft recently highlighted that Florence v1.0 now tops several industry benchmarks, check out some of them below.
- TextCaps Challenge 2021
- Kinetics-400 action classification
- Kinetics-600 action classification
- OK-VQA Leaderboard
- nocaps Leaderboard