I would like to create a program which takes an image of a scene (for example an image of a bathroom) and the program is able to identify the objects and their position in the image and recreate the scene in 3D.
After recognizing the objects in the image, the program will place 3D models (matching placeholder 3D models which are stored in the program) of the objects in a 3D scene. The program would also be able to get an idea of the object's position from the image and place the objects in the 3D scene according to this position.
Can anyone refer me to any research papers or already existing programs which explore this idea. (Would also appreciate if these would be on the simple side as my knowledge on the area is limited.)