Corner Pin Node

The Corner Pin node is used to both distort and "fly" one image over another through a series of frames.

The Corner Pin node accepts one or two inputs: the foreground image and an optional background image used in the work monitor to establish size and provide a visual reference for the motion.

The corner positions can be controlled interactively in the work monitor, which features an adjustable bounding box for manipulating the foreground image interactively.

The interactive controls in the work monitor are tied to the four Corner parameters on the node panel and their range of motion can be constrained, if you so specify, by the Shape Limits menu.

All four corners can be animated, and most commonly the Corner Pin node uses position data from a Track node to control the corner values.

Using Track Data in Conjunction with the Corner Pin Node

You can use expressions to access data generated in a Track Node to control the position of each corner. Enter the following expressions in the x and y parameter fields of the Corner parameters:

Corner 1: track("track1", XV, 1, $F) track ("track1", YV, 1, $F)
Corner 2: track("track1", XV, 2, $F) track ("track1", YV, 2, $F)
Corner 3: track("track1", XV, 3, $F) track ("track1", YV, 3, $F)
Corner 4: track("track1", XV, 4, $F) track ("track1", YV, 4, $F)

These expressions stipulate

  • that the data should be accessed from a node called "track1." (You should use the exact name, in quotes, of the actual Track node that will be accessed.)
  • whether the x or y value (XV or YV) should be used.
  • which track point to use (1, 2, 3, or 4).
  • that the data for the current frame number ($F) should be used.

Tip:
When placing the four track points in the Track node, position them so that their numbers correspond to the Corner parameter numbers in the Corner Pin node: place Track Point 1 in the lower-left corner of the object to be tracked, since the corresponding corner in the Corner Pin node is Corner 1; place Track Point 2 in the lower-right corner (Corner 2), and so on. This will ensure that you reference the correct point numbers in the above expressions.

Corner Pin Parameter Tab

Work Monitor / At Resolution

First, select a display resolution for the work monitor from the At Resolution popup menu. Then click the plus icon to display the work monitor for this node.

Transparency

The Corner Pin node enables you to input a background image for reference. The Transparency parameter is used to specify the level of transparency of the top input over the bottom input as it is viewed in a temporary composite in the Corner Pin work monitor. You set the transparency value of the top input by keying in a numerical value between 0 (transparent) and 1 (opaque) or by using the mouse to set the tuner to the desired value.

Use Image Alpha with Transparency

When this option is checked, the alpha channel value of each pixel of the primary input is multiplied by the Transparency parameter value to govern the overall transparency of the temporary composite image in the work monitor. (This is assuming that you have used the bottom input connector to the Corner Pin node to input a background image for reference.)

Shape Limits Menu

The Shape Limits parameter enables you to set certain constraints on the movement of the image being pinned:

Move Corners Independently

This mode, which is the default, is unconstrained. You can drag the corners of your image freely to move, scale, rotate, or distort their shape.

Right Angles with Rotation

In this mode, you can move, squeeze, and stretch your image and also rotate it against the background frame. The opposite sides of your image, however, will remain parallel.

Maintain Right Angles

In this mode, the borders of the image always remain parallel to the borders of the frames. You can move, squeeze, and stretch your image, but it will remain perfectly aligned.

Corners 1-4

You can key a pair of values into each of the Corner 1 through Corner 4 parameters. Each parameter accepts two values between 0 and 1.

  • Corner 1 represents the lower-left corner of the cropping box.
  • Corner 2 represents the lower-right corner of the cropping box.
  • Corner 3 represents the upper-right corner of the cropping box.
  • Corner 4 represents the upper-left corner of the cropping box.

Pixel coordinates are produced using the following computation, using Corner 1 as an example.. If we define the Corner 1 values as X1 and Y1:

  • X1 multiplied by the width of the source image will generate the X pixel coordinate.
  • Y1 multiplied by the height of the source image will generate the Y pixel coordinate.

Filtering Parameter Tab

The Filtering tab enables you to select an appropriate filtering option for your imagery from the Filtering popup menu.

No Filtering

This option will give you the fastest result and may be adequate for previews; however, it is not recommended for final output.

Super Sampling

Super sampling is adequate if you are performing only a translation. But this form of filtering degrades the image when scaling to a significant degree. The associated AA Level parameter allows you to set the level of anti-aliasing.

AA Level

The AA Level parameter becomes active if Super Sampling is chosen from the Filtering menu. The AA Level controls the level of anti-aliasing in the image. Anti-aliasing is a filtering technique that creates the appearance of smooth lines and edges on a raster image display.

Select one of six levels of anti-aliasing using the popup menu. Level 1 provides the least amount of smoothing; level 6 provides the maximum. Level 3 is the default. The higher the level of anti-aliasing, the longer the node will take to complete its processing, but the sharper the image will be. You may want to use a low level of anti-aliasing to experiment, and then reprocess with a higher AA Level setting.

Two Pass Linear

Two pass linear and Mitchell filtering are both better choices than super sampling for major scaling or rotation operations. Two pass linear is preferable when scaling an image, as long as the image is not very noisy, because it is a bit faster than Mitchell.

Two Pass Mitchell

Two pass Mitchell is an excellent low pass filter for noisy images. It is likely to be the best choice for Jpegs or images with a lot of fine lines.

Mitchell B and C Parameters

You can make adjustments using the Mitchell B and C parameters, which become active when Two Pass Mitchell filtering is selected.

The default values, 0.333 for both parameters, fall within a recommended range of roughly 0.25 to 0.5. Since the quality of the result must be judged based on your imagery and intentions, the best course may be to experiment with these settings.

As a guideline, note that values above 0.5 for the Mitchell B parameter may produce unnecessary blurring. Values above 0.5 for the Mitchell C parameter may result in excessive ringing (rippling patterns). When both parameters are set at values above 0.6 or below 0.2, anisotropic artifacts may appear.

Two Pass Lanczos

The Lanczos filter provides a good compromise between sharpness, ringing and aliasing reduction. It is best used on images which are reduced or scaled down.




Table of Contents | Index



Copyright 1996-1999 by Silicon Grail Corporation