The arrival of the App Store gave rise to a large amount of new software that during the first years of its existence largely defined the novelty of having a centralized site where to buy and download applications. The first games appeared, the first applications characteristic of the use of the gyroscope and when the social networks became strong on the platform, the first uses dedicated to them.
One of the most widespread uses is the use of filter applications . In this world of the pocket image, the devices became perfect for this type of software for two main reasons: to carry at any time a camera and the great power to process images despite being such small devices.
Social networks popularized much more – if possible – the use of filters, even introducing them into every major social application as one more functionality. Instagram, Facebook, Twitter… all of them even put names to each effect in order to make them more personal, the most important goal for which we would like to apply filters: to make it more ours, more personal, more different. And during all these years, many of them have come up with very good results… although Prism , is something else.
Prism, here comes artificial intelligence
We could pigeonhole Prism as another application to apply filters, but that would not be entirely correct, not even fair. The small development team located in Moscow probably thought that the world of filters was too saturated , because they all started with the same premise: apply a layer that graphically alters the image and that can be combined with each other.
Prisma uses an AI algorithm similar to that used by Google to detect human shapes and objects
Prisma starts from the opposite side: not only the effects cannot be combined – in an app like Prisma it would not make sense – at least without resorting to “tricks”, but the most remarkable thing about this application and what makes it really remarkable is the use of artificial intelligence to detect patterns and shapes . This allows for differences in people, objects and image backgrounds, for example, so that the application of the filter algorithm is more accurate and faithful.
The way they do it is not usual either: the image is processed in the Prisma server , not in the mobile, so it does not depend on the processing capacity of the device, but on the availability of the server. Trying the application, I found some times where the service was not available due to the large amount of users connected, but I could use it normally, with a more than acceptable speed ( just 15 seconds to apply and download each pattern ).
Prisma (gratuito, App Store)
More than applying filters, Prisma creates new images with algorithms based on recognized styles , so from a photograph we can see how it would look and we realize how artificial intelligence is being applied when it comes to recognizing objects and people: defined and differentiated profiles of objects and background. Traces that enhance the main figure more than normalize the whole image with a layer with several filters, as we are used to.
It is a very visual application: few steps and clear functions so that what is important is our creation
The application also has a simple and direct graphic interface : one part to capture the image, and another part to apply the filter and share it. No more confusion, and with a more than correct speed – if the server is supporting the traffic on the spot. It has limitations, such as the size and proportion of the images that it can process that does not yet support video processing ( although the authors claim that this is a feature that will come in the future ).
Is prism the perfect tool to create an original image and away from conventional filters to share on our social networks? No doubt, since it will be much less recognizable than the modifications we make with other similar ones, although undoubtedly the most powerful thing about the free app ( , by the way ) is its process entirely on the server and the power of its artificial intelligence.
App of the Week : Prisma uses image processing using artificial intelligence algorithms to achieve more personal filters, away from traditional modification layers. An original and powerful idea that brings a breath of fresh air to a type of application.