Microsoft already had an AI service that can generate captions for images automatically. Today, Microsoft announced that it has achieved human parity in image captioning on the novel object captioning at scale (nocaps) benchmark. With this innovative technology, the image captions generated by AI are more accurate than what was previously possible. This is not a research announcement. This new improved image captioning capability is now available to all customers via Azure Cognitive Services.
Microsoft will also be using the new image captioning tech in Seeing AI app. It will start rolling out later this year in Microsoft Word and Outlook, for Windows and Mac, and PowerPoint for Windows, Mac and web.
“Microsoft researchers have built an artificial intelligence system that can generate captions for images that are, in many cases, more accurate than the descriptions people write. The breakthrough in a benchmark challenge is a milestone in Microsoft’s push to make its products and services inclusive and accessible to all users,” wrote Microsoft.