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…
Google’s AI is scary good at depicting what’s in your photos
No comments:
Post a Comment