Learn more about the three camera components: camera eye, camera view, and camera up vector.
The camera eye can be considered the camera unit itself, which travels along the motion path to which it is assigned. The motion path determines the position of the camera at any given time in relation to the scene. For example, the design of the motion path lets you move the camera closer to or further away from an object in the scene. To visualize, think of yourself walking along with a camera taking pictures: the route you follow would be the motion path.
The camera view can be considered the focus point of the camera — where the camera is looking at any given time. If the camera view is not assigned to a motion path of its own, then the view is always directly in front of the camera. Think of yourself walking through a scene without ever pivoting your head, your view is always directly ahead of your body. By assigning the view to a motion path of its own, you can change the view point of the camera at any time in relation to the camera position.
The camera up vector can be considered the current angle of the camera at any given time in relation to the camera eye. The camera up vector is the direction from the camera’s eye to the camera’s up. If the camera’s up is not assigned a motion path of its own, the camera remains parallel to the path that the eye has been assigned to at all times. By assigning the camera up to an independent motion path, the camera can be pivoted to any angle up to 360 degrees. This can be likened to a camera on a tripod except that the tripod could pivot a full 360 degrees.
At least one curve is needed to use Autofly. If only one curve is used, it must be the motion path for the camera eye. The camera view then remains directly in front of the camera throughout the animation.This might be exactly what you want. If not, and you want to control the camera view throughout the animation, you need at least two motion path curves.