One of the most remarkable facts of the human visual system is that it rapidly and accurately understands the characteristics of the complex visual world - the relative depth with respect to different objects in the scene, occluded objects in the scene, etc. due to prior experience and knowledge about the scene. The various types of tasks related to understanding what we see in a visual scene is called `visual recognition'. In the field of computer vision, visual recognition has had some great success in recent years and is still developing further. To make computers understand the various characteristics of the complex world is a challenging task in computer vision systems. New approaches came into existence which predicted depth from single images and global image features were also taken into consideration. Thus due to the accuracy of results the importance of global image features were understood. This paper deals with a discussion of a few well known approaches leading to creation of an approximate 3D model from a single still image.