Applications like Facebook, Snapchat, Pinterest, and Instagram are now indispensable for people at different ages, genders, or nationalities. Social Media has been sprawling into our lives, through our smartphones, becoming a part of our daily routine. It has substituted the need for traditional Newspapers for many of us. But have you ever wondered about the technology behind smartphone cameras, which take pictures for such apps? As the bar of the competition is raising by the day, the development of picture editing software has taken a significant leap forward. This can be seen in smartphone cameras with incorporated HDR settings, as well as the variety of filters offered by apps like Instagram.
Image via PDF uploaded by researcher Jon Barron
According to MIT News, a collaboration between Google and MIT’s Computer Science and Artificial Intelligence Laboratory resulted in a much-awaited algorithm that takes image-capturing to a whole new dimension. The algorithm allows the automatic fine-tuning of the image quality in almost no time. Moreover, you can see a preview of the image even before shooting it.
The algorithm was presented at the Digital Graphics Conference “Siggraph”, last week. Image Contrast, saturation, color balance, and brightness are all subjects of improvement in this algorithm. Such developed controls help us, non-professional photographers, to achieve perfection and produce top-quality yet natural-looking photographs. The new system is much faster than manually editing the photograph after shooting it. The improved camera performance involves top-notch filters which can make changes to the picture uniformly or according to simple data, like the brightness of pixels.
The concept is based on an automated learning system which has surveyed thousands of images to recognize how they were edited by skilled photographers. Later, the information enters an algorithm permitting photo-editing in a fraction of a second, giving us a chance to see the changes before we take the shot. To learn more about the algorithm, watch the following video.
You can also get further details by checking the article on the MIT News forum.
