Deep learning algorithms created by researchers at the Indian Institute of Technology (IIT) Madras and Northwestern University in the United States can considerably improve depth perception and 3D effects in videos captured with smartphone cameras. Officials claim that such algorithms will prevent “flat” mobile phone photographs and give them a true 3D sense. One of the most significant advantages of the created algorithm is that it eliminates the need for expensive equipment or a variety of lenses to capture videos with depth.
According to Kaushik Mitra, assistant professor at IIT Madras’ Department of Electrical Engineering “Photographs and films are taken with smartphone cameras have a flat, two-dimensional aspect, which is a popular complaint among pro and amateur photographers. Apart from the flat appearance, several 3D characteristics, such as the Bokeh Effect, or the artistic blurring of the background, which is simple with a DSLR camera, are difficult to achieve with a smartphone camera.”
In addition, he said, “While a few mid-and high-end smartphone cameras are already designed to incorporate similar effects in still images, particularly in portrait mode, it is currently not viable to reproduce them in smartphone movies.” To give the illusion of depth, powerful professional cameras gather information about both the intensity and direction of light in a scene, which is known as Light Field (LF), according to Mitra.
According to him, “An array of micro lenses is put between the camera’s main lens and the camera sensor to accomplish LF capture. Due to space limits, many micro lenses cannot be used on mobile phones. However, algorithms are now being designed to post-process the images collected by current mobile sensors.”
“For picture alteration, artificial intelligence and machine learning techniques are applied. Our team investigated the problem and developed a deep learning technique that converts stereo photos taken with a smartphone into LF images.” he added.
ALSO READ: Apple IPhone Users Reportedly Facing Battery Draining Issue After Updating To IOS 15.4 | OyPrice
The findings were presented in the ‘Proceedings of the 2021 International Conference on Computer Vision (ICCV)’. Kaushik Mitra explained “The algorithm begins by simultaneously capturing two movies known as a stereo pair with the two neighboring cameras found in many modern smartphones. Deep learning models are used to take these stereo pairs through a series of phases. The stereo pairs are turned into a 7X7 grid of photos, simulating a 7X7 array of cameras, and the LF image is created.”
Further he added, “One of the most significant advantages of our team’s algorithm is that it eliminates the need for expensive equipment or a variety of lenses to create videos with depth. With a smartphone equipped with a dual-camera system, Bokeh and other attractive 3D effects can be generated. In addition to giving depth, our technique allows us to see the same movie from any of the 7×7 angles.
Join us on Telegram, Facebook, or Twitter and stay up to date with the latest happenings in the tech industry.