The search engine giant is popular all over the world for its innovations in various fields of technology on a regular basis. It can be rightly said that even a fortnight does not pass without the news of Google’s latest innovation popping up. It seems that the tech giant has pledged to take technology on a different level altogether. Especially in the case of cameras, the demand for better and better technology is increasing. Be it digital cameras or the casual phone ones, everyone wants the best result possible. In order to fulfill this demand, the tech giant has collaborated with MIT Research and created the latest Google photo editing software.
Currently, all cameras work on a common concept. This concept treats the captured image as the raw material which has to be worked upon in order to enhance the look of the image. But what if the enhancement takes place even before you click the image? This is what this Google photo editing software aims to do. Google partnered with MIT Research to create software which can retouch images in real time before they are clicked. This in itself gives us an idea about the speed of the software as it instantly shows the enhanced changed.
The highlight of this Google photo editing software is that it has slimmed down the algorithms usually used for such purposes.
This Google photo editing software works on algorithms
This photo editing software is mainly on the technical concept of computational photography. This means that the software relies on machine learning and algorithms to improve the quality of your images. The researchers at MIT trained neural networks on a data set holding more than 5,000 images to create this software. Adobe and MIT have put together this image collection. Five different photographers have retouched every single picture in the collection. The algorithms will then use these images as a reference point to learn about what kind of improvements to make.
The highlight of this Google photo editing software is that it has slimmed down the algorithms usually used for such purposes. This makes the software lighter yet efficient enough so that it can run smoothly without any lag. The size of the software is almost as much as a single digital image. This small software is capable of processing images in a range of styles. The software can further train itself on other images to expand the retouches it can carry out. One can find its official announcement on the official MIT blog.
