You will need a good computer graphics program. Photoshop or PaintShop Pro work. The program must be able to split each colour photo into red, blue, and green 'channels', and then recombine selected channels into new images. You also should be familiar with how to paste a transparent image, in case you need to do some adjustments on how the pictures line up. For each image, you will have to take two pictures. Find something that won't move, and stand about a metre away. You are imitating what each of your eyes sees, so you'll need to take the first photo from slightly left of center (where your left eye is), and then the second from about 4 inches (10 cm) to the right. For each photo, point the camera directly at the subject. But be careful not to move the camera up or down, or forwards or backwards, between shots, and if you have a live subject, make sure it doesn't move at all. For objects or people farther away, increase the distance slightly between pictures. Scenery in the far distance might require a distance of up to a metre between pictures ... you'll have to experiment. Load the two pictures into your graphics program. Make sure they are exactly the same size. (If they aren't, you must retake the pictures). Now find the menu that lets you split and combine RGB channels. Select the image that was taken from the right, and split it into three channels (red, green, blue). These new pictures will look grey ... that's OK. Now do the same for the left side picture. When you're done, you'll have six new pictures ... but we're only going to use three. The final step is to use the menu to combine channels. You'll want to combine the red channel from the right image, and the blue and green channels from the left image. This will give you a brand new picture that looks like the images in our gallery. If you've done everything correctly, you will see images that seem to overlap. Check it with the glasses. You need to make sure the overlap is a millimetre or less (for close subjects), and that the overlapping images are on the same horizontal level. If this isn't true, you'll need to start over, or do the following. Close all the six channel images, and start all over, with your two left/right images. Crop one of the pictures slightly on the left and/or top to make it line up with the other one. Your graphics program should let you paste images in transparent format; if you copy the cropped one and paste it, transparently, onto the other, you can see in advance whether they are going to line up properly or not (same horizontal level, and about 1 mm difference, left to right). If they don't, go back and recrop. When they do, paste the cropped picture onto a blank image that's the same size as the unaltered one. Don't worry about borders ... they can be cropped out when you're done. Now split each into RGB channels again, recombine the right red and blue & green left channels, and check this new image for 3D effect. You may need to do some experimenting to get a good 3D effect. Keep your glasses handy, and use them to check the combined photo. The best way to avoid problems is to make sure the subject does not move at all, and that the camera is moved in a perfect horizontal line. |