Google’s AI is scary good at depicting what’s in your photos


Watch out IBM Watson, Google has its own kickass ‘Show and Tell’ AI and it’s getting pretty damn good at depicting what it sees in photos – and now everyone can use it. Today, the tech giant announced it’s open-sourcing its automatic image-captioning algorithm as a model in TensorFlow for everyone to use. This means anyone can now train the algorithm to recognize various objects in photos with up to 93.9 percent accuracy – a significant improvement to the 89.6 percent that the company touted when the project initially launched back in 2014. Training ‘Show and Tell’ requires feeding it hundreds of thousands of human-captioned images that the machine then uses and re-uses when…

This story continues at The Next Web


RSS-4

Related Posts:

Admin