Artificial intelligence to put an end to failed selfies
- Author: Carolyn Briggs Apr 07, 2017,
Apr 07, 2017, 23:33
It is sophisticated AI machine learning that runs intense checks on an image before applying changes.
Last autumn, Adobe announced Sensei, which is a framework of intelligent services created to enhance digital experiences.
Powered by Adobe Sensei, a sophisticated technology that combines artificial intelligence and machine learning, the retouching enhancements provided by the new software make it a fully-fledged magic wand.
Adobe is using its Sensei platform on its Creative Cloud tools and its Marketing and Document Cloud services. Adobe promises it can transform a typical selfie into a "flattering portrait with a pleasing depth-of-field effect that can also replicate the style of another portrait photo".
Adobe made the code for this "Deep Photo Style Transfer" open source and it is available on GitHub for interested developers. While they haven't released any real details other than the demo shown in the video, it's still fun to watch. Last but not least, users also have the option of cutting and pasting photo styles from one image to another. He was able to simply tap on a photo he liked, and Adobe's app was able to copy and paste that photo's style and apply it to the user's own selfie. "Great portrait photography requires the right perspective, equipment, and editing expertise", Adobe said.
Adobe today shared a video from its Adobe Research team, demonstrating what future selfies could look like using artificial intelligence and deep learning technology.
It shows a dude taking a bad, embarrassing selfie and improving it using tools that apply artificial depth of field, tweak the perspective so it's not clear it was taken from so close to the face, and steal aesthetics from other portraits of other dudes. This is the first time that Adobe demoed its capabilities on a smartphone, as pointed out by TechCrunch.