Berkeley and Adobe

The researchers from the University of Berkeley and Adobe have demonstrated about a photo editing tool that enables sophisticated photo modification using target images from the source image. The final result is a realistically modified image that has been machine modified to match the target image without extensive ‘natural’ user editing.

Adobe’s excellent software has firmly established the company as one of the most influential sources of innovation in image manipulation over the years. It seems the Photoshop-maker intends to keep things this way. According to the technology published recently, the researchers have claimed a new method that generates “photo-realistic samples” entirely based on a few brush strokes from the user which is interesting too.

The deep learning tool also comes with a ‘Generative Image Manipulation’ feature that automatically transforms the color and shape of one image to resemble like another. For instance, drawing a general shape over a photo of a bag causes the software to adjust the bag’s size automatically to match the sketched shape without any loss in its realistic nature. The software can also generate images based on boorish user ‘scribbles’. There is no requirement for artistic or artificial talent.

This editing tool applies alterations to the original photo by considering the input and using the intelligent generative algorithm to produce a more realistic output. In the absence of a source image, the tool will generate results from scratch.

Watch the below video to get a clear idea about how the tool works.

You May Also Like : How To Resize And Enlarge Images Without Disturbing Quality