Ray tracing rendering depth (Depth of field) based on the effect achieved

Offline rendering graphics used when a perspective camera model in accordance with the principles of the pinhole imaging, which is usually achieved emitted from the eye to the ray on the imaging plane, and the results of the ray trace of this sample as a result of the imaging plane corresponding to the intersection . That is:

the-lens image captured in real effective depth (Depth of field), such as:

Clearly, the above figure earphone is placed near the focal plane of the camera, very clear image; background scenery from the focal plane than far, showing a blurred effect. Offline rendering, the effect of this will help strengthen the authenticity and layered rendering results. But before mentioned camera model is too simple, to get the depth of field, you need to be more accurate camera model.

The figure is a schematic view about the depth of field, terrifying seem complicated given network. But a little analysis can know that the scene is not on the focal plane appears the most fundamental reason is precisely the vague concept of focal plane itself, as shown below -

you can see if adding a convex lens in the original perspective camera model, then imaging process will no longer be "a" of light through the hole reaches the point on the imaging plane, but the lot of light to a point on the imaging plane through the lens.

Obviously, if there is exactly one object on the focal plane, the emitted light from a point above by the convex lens will converge to a point on the imaging plane, then a clear image is obtained. However, when the object is not in the focal plane, the rays converge to a point on the imaging plane will not all from the point on the object, but from a small area on the object, which formed a blurring effect, the final is seeing the depth of field.