“We live in a 3D world but when we click a picture it comes in 2D,” says Tianfu Wu. Recently researchers have developed a new technology called MonoCon. The MonoCon improves the ability of AI programs to identify the 3D objects and how those objects relate to each other in space using 2D images.

The Artificial Intelligence programs receive input from the cameras. So if we want the AIs to understand and interact with the world better. It is important that they are able to interpret 2D images and tell about 3D space. The main challenge is to make AIs accurately recognize 3D objects such as people or cars in 2D images and place them in the right space. It will work as an important application for not only autonomous vehicles but also for the manufacturing and robotics industries. 

The MonoCon is a new advanced technology that is capable of identifying 3D objects in 2D images and placing them in a “bounding box”. The bounding box informs the AI of the outermost edges of the relevant object. The MonoCon was tested by the researchers using the benchmark data set called KITTI. The aim of building MonoCon was to help the AI programs extract 3D data from 2D images.