PHOTOGRAPHY 3D

We present an approach to convert a small portion of a light field with extracted depth information into a cinematic effect with simulated, smooth camera motion that exhibits a sense of 3D parallax. We develop a taxonomy of the cinematic conventions of these effects, distilled from observations of documentary film footage and organized by the number of subjects of interest in the scene. We present an automatic, content-aware approach to apply these cinematic conventions to an input light field. A face detector identifies subjects of interest. We then optimize for a camera path that conforms to a cinematic convention, maximizes apparent parallax, and avoids missing information in the input. We describe a GPU-accelerated, temporally coherent rendering algorithm that allows users to create more complex camera moves interactively, while experimenting with effects such as focal length, depth of field, and selective, depth-based desaturation or brightening. We evaluate and demonstrate our approach on a wide variety of scenes and present a user study that compares our 3D cinematic effects to their 2D counterparts.

 How to take 3D photos

A couple of weeks ago, I wrote a tutorial on how to see cross-eye 3D stereo photos on your computer screen. Today I’m going to tell you how to take your own 3D photos.
The process is really very simple, and the basics can be explained in less than a minute, but to become good at taking and presenting 3D photos take a bit more time, and it’s something that really develops with practice. I hope you’ll take what you learn here and get out and get lots of practice taking 3D photos.
The typical and easiest single-camera 3D photography technique is commonly called the “cha-cha” technique, for reasons which will soon become obvious. Here’s the technique in a nutshell:

Taking the Photos

• Always take photos in portrait orientation
• Set the camera to full manual and choose the correct exposure and focus
• Put your feet square on to the subject of your 3d photo
• Put your weight onto your right foot, without lifting your left
• Take the first photo
• Put your weight onto your left foot without raising your right
• Unless your subject is very close, you don’t need to turn your camera to keep it in the centre
• Take the second photo

And you’re done! As simple as that. Simply by shifting your weight from one foot to the other, you move your viewpoint by several centimetres, sufficient to get a 3D effect. The “cha-cha” name refers to the side to side sway you do when taking the photo.

proccesing
Now you’ve got two photos, but you need to make one. I use a fantastic piece of free software called StereoPhoto Maker. It’s not the prettiest software, but it does a fantastic job. It can be used in conjunction with a plugin called AutoPano, that can analise the two images and automatically correct for many of the problems that can come from shooting two separate images. This includes tilting and twisting, moving forward or back between shots, and the “keystone distortion” that occurs when you turn the camera to centre the subject for close 3D photos.

Step by step

• Drag both photos onto the StereoPhoto Maker shortcut icon
• Zoom out a bit with your mouse’s scroll wheel if you need to to make it easier to fuse the pair into 3D with the cross-eye technique
• If the 3D effect seems reversed, click the swap button to swap the images correctly for a crossed eye view
• Click on the auto align button to use autopano to correct for any distortions
• Click on the Easy Adjustment button to fix the 3D images position relative to the 3D window (see below for more information)
• I find adding a border helps, so if you like, in the menu go to View – Border Options
• Check “Show Border” and adjust the border settings to your liking
• Save the image by clicking in the menu File – Save Stereo Image
• Enjoy your new 3d photo!

The 3D “Window”

The edge of the image is more than just the boundaries of the 3D photograph. In a 3D photo, it is also a “hole” into which you look and through which 3D subjects can appear. A good way to think of the edge of the image is as a literal window in your computer screen. This is one of the reasons why I find a border around both parts of the 3D image helps me, it more clearly defines the edge of the 3D window.
Just like a real window, you expect to look through it, and rarely do you expect things seen beyond it to come back through it at you. One nasty optical illusion that can happen with 3D photos is when part of the 3D subject “touches” the window, or worse, appears to overlap it. Have a look at the two examples below:

The subject is placed too far forward in the 3D window, and appears to overlap the window’s edge, creating an uncomfortable optical illusion. This is most obvious on the left of this stereo photoHow to View

Here the photo has been corrected, pushing the 3D subject back in 3D space beyond the window frame.

How to View

Now that you’ve got a whole bunch of awesome 3D photos, it’s time to share them around.

If you’d like to share your new 3D masterpieces, and I encourage you to do so, there’s a Stereophotography Flickr group, and one on Yahoo too. I’d love you to put links to your 3D photos in the comments to this post, so I can see how you’ve done! I’m relatively new to 3D photography, so I’m sure there’s much you can teach me too!

Advantages

• You don’t need expensive specialist 3D equipment
• Each eye’s image is captured on a full frame, so the resulting 3D image can be very high resolution
• There is no blurring or ghosting at the edge of the frame, which can be seen in many “beam splitter” attachments where the two views join
• You can take a 3D photo with any lens in your SLR kit, including macro, for extremely close 3D photos
• Many 3D attachments have very limited control over focus and aperture, with the cha-cha technique you have complete control over the settings
• You can do this with any camera, if you forget to take your 3D attachment or camera, you can still take 3D photos this way

Disadvantages

• The most obvious and critical shortcoming is that this method only works with still object that don’t move between shots
• Any movement between the two shots will cause a distracting 3D error, so people, animals and even trees in a light breeze will be difficult or impossible to shoot
• You need to take two photos for every 3D image, which takes twice as long, and uses twice as much space
• It is easy to introduce errors such as twisting or tilting the camera between shots which can cause distracting artifacts
• You need to shoot with manual settings so that there’s no accidental variation in exposure or focus