The developer community were in for a great news, and this time it was from the Search Engine giant Google. Google has decided to release a new set of APIs named as TensorFlow. It is an object detection API which will allow developers to identify objects inside images. Sources say that the organization has put lot of hard work to get these APIs up and developed almost perfect models.

The models incorporated in these API files are neural networks based on inceptions; and these models are totally streamlined. The beauty of these streamlined models is that they can work on machines that are less complicated. What they mean by less complicated machines is smartphones. It is better to have these models device agnostic otherwise, for every data transfer it needs to be called from the cloud serve, which can cause potential latency. Over and above this you need an uninterrupted internet connection for a smooth functioning.

Developers have the flexibility of choosing from a variety of models, having a multitude of parameters and computations for processing a single image. Developers who are planning to train their own developed models, can use the scripts provided by Google both locally as well as on cloud platform. The idea behind this initiative of Google is to push its various tools for artificial intelligence to a larger audience, which will further stimulate the creation of many more such models and advanced systems; by allowing developers to work on Google toolkits and source codes and build a model on top of Google’s framework. Developers get the opportunity to tradeoff between accuracy of the application and battery power.

It is reported that these capabilities of TensorFlow will prove to be advantageous significantly, as it will be a notch above the other frameworks; as this will enable people using machine learning frameworks to mind share. Since this model is released as an object detection system which can run on mobile devices, it turns out to be a key USP for Google. This is based on MobileNets image recognition model which the tech giant open sourced earlier. Google has also come out with a lighter version of this framework, called TensorFlow Lite.