1. Segment Button

FG/BG panel: Use this to identify the colors in the foreground (left mouse button) and background (right mouse button). If the foreground-background colors are different, often you can just make one sweep across a representative area on each and that is enough.

Then click Segment.

That causes the Foreground panel to come up, where you can work on fine detail like fly-away hair. Now, the cursor is like an intelligent paint brush. If you use the left mouse button, it is only active on things that the software thinks are in the foreground. If you use the right mouse button, it is only active on things that the program thinks are in the background. Even if you run the brush across the background with the left mouse button, if the colors there are different, it won't have any effect there.

If you need to see the reverse image, switch from the Foreground panel to the Background panel.

The main reason for the Foreground and Background panels is to let you fine-tune details like tree branches or fly-away hair. Using the left mouse button, you can sweep the cursor back and forth over these on the background, and pull them into the foreground. (It is impossible to do this in Photoshop.) Similarly, if you can see the background through strands of hair or tree branches, you can use the right mouse and sweep over that whole area, and the background can be separated from the hair and set in the background.

Then you can click Done, and generate the 3D model.

If you make a mistake, or if the same color is in part of the foreground and background, you may cause something unintended to happen. That doesn't happen if there are different colors in the foreground than background. But you can always undo by going back to the first FG/BG panel. You can mark that area with the left or right mouse, adding to what you had previously marked. Then click segment again. The fine detail can then be marked on to the new image.

2. Settings Panel

Configuration Panel: The settings in the configuration panel will stay for repeated uses. You need to press Apply when adjusting them, to cause the 3D model to be rebuilt.

Output Size: This is one setting that cannot be changed after the image processing starts. You may want to just leave this at a size that gives you a suitable image size for your printing and interlacing.  But if you decide to change it for a new image, it must be changed as the first step before the image processing.

Lateral Translation vs. Rotation: 3D viewing is generally most effective with Rotation around the foreground object (the same as a real object in your hand rotates when you tilt it back and forth). However, if you need sharper printing or your goal is stereoscopic (2-frame) pop-out, you may want to try Lateral Translation. As a late addition to the interface based on a customer request, there is still more learning to do on optimal usage of Lateral Translation with no toe-in.  For most purposes you should use Rotation; Lateral Translation is just experimental at this point. If using Lateral Translation, you may find that a narrow Camera Angle (e.g. 20 degrees) needs to be increased, if there is not enough relative motion (parallax) between the foreground and background.

Image Size: Small initial images are fine, since the final interlaced image will be larger.  At least for ink jet printing, even images as small as 500 x 380 have enough resolution to give you more than enough interlaced strips. This also has the advantage that the 3D wireframe processing is faster with smaller images.

Orientation: Ink jet printers may have different resolution in landscape and portrait printing.  The simplest approach is to use 8x10 lenticular plastic with vertical ridges, and print in portrait orientation.

Background Movement: Rather than a homogeneous color, having patterns or details moving in the background provides a good depth cue.  To get the clearest background image, you want a balance between having the background move (which is a good depth cue), versus having it move too much in big steps (which the eyes can't fuse, creating ghosting and blur). Setting the Camera Angle as low as possible makes the background appear flatter, which will help. The background will also be clearer with less rotation (e.g. 20 degrees or under). Or, use more rotation but minimize the background depth to maintain the clarity of the background.

Foreground Movement: Just as with the background, if the foreground moves too far or in too large of steps, it will appear ghosted, or less sharp, or could seem different to each eye. This sample picture shows what the interlaced image looks like (in reduced size) before the lenticular lamination. But the rotation is larger at the outer edges, causing the bride's face to be less sharp than the groom's.  The two red lines show the difference in the widths of the eyes because of the rotation.  This is easily solved by reducing the amount of rotation, or possibly also by moving the camera back or changing the camera angle.   

Number of Frames:  Print as many as you are able to interlace, given the resolution of your printer.  This makes the movement seem smoother, if the steps are not too large from one frame to the next. Ink jet printers can commonly print at 600 or 1200 dots per inch. To know how many frames you can print for interlacing, use this formula:

Dots per inch / LPI = Number of Frames

If your printer has a resolution of 1200 horizontally, and we recommend using 40 lenses per inch, then you can print up to 30 frames for interlacing. Note that although people love the new Unlimited Sweet Spot (tm) option, you need to specify fewer frames when using it. 

Hiding the Background Shadow: If you keep the same background from the original photo, there will be an occluded area behind the foreground that the camera could not see. This is not an issue if you use a different background image.  But when keeping the same background, here are the settings that will minimize the size of the background hole for viewing from all angles. You can minimize the visiblity of the occluded shadow by moving the Camera Closer, by lessening the amount of rotation, by minimizing the backdrop distance, and by increasing the Camera Angle.

This is a brand new technology, so please feel free to let us know of any new techniques that you think others should know about.

Software Overview  | Software Instructions | Tutorial | Download | FAQs