I think you agree that Google Pixel cameras are one of the best cameras. Google used top-notch camera sensors inside. But what makes it more magical? It’s own software. Yes, Google Pixel device software adds some magic to the photos captured by its camera which looks really amazing.

This is Awesome!

The Google Pixel camera is using the Semantic image segmentation. Now, what that means? It is a method to separate things which are covered in the image. When anyone captures an image in the device, this technology labels everything like “sky”, “person”, “dog”, “road”, etc. to every pixel which is present in the image. It helps to enable some application for what user is asking like synthetic shallow depth-of-field. Google Pixel 2 and Pixel 2 XL comes with the only single camera placement on the back side. There is no need for the second camera. All work does their super smart camera technology. Because of this, Pixel 2 cameras can capture best photos in portrait mode.

Google is Happy

As per mentioned in their official Research blog post, Google released this technology as an open source project. Now any developer can use this technology in their camera application and also developer can add some extra cool stuff and features in the camera. So definitely they will now be able to any modification. Portrait mode is the one example of what everyone might install the camera applications on their Android devices.

Mentioned in their post

This release includes DeepLab-v3+ models built on top of a powerful convolutional neural network (CNN) backbone architecture [2, 3] for the most accurate results, intended for server-side deployment. As part of this release, we are additionally sharing our Tensorflow model training and evaluation code, as well as models already pre-trained on the Pascal VOC 2012 and Cityscapes benchmark semantic segmentation tasks.

With this Google says that they are happy to make it open source. Publicly sharing their system in the whole Android community will make easy for all developers to make it even better to use them on the Android device. Also, it can be used for the further applications as a part of new technology.

Google Research Blog