Understanding 3D

Bookmark and Share

Mon, 03/08/2010 - 19:00 -- Nick Dager

By Matthew DeJohn and Michael Karagosian There is a shifting tide in the film industry and you can only see it if you wear the right glasses: 3D glasses. At the time of this writing there are nine thousand 3D capable theatre screens worldwide and scores more being installed. The home entertainment market is set to release dozens of 3D TV and Blu-Ray models. Avatar made 3D history and more than 25 3D movies are slated for release in 2010. It's clear that 3D is becoming more than just a trend. So for anyone involved in the movie industry it's a good time to understand what 3D is all about. How does the illusion work?  How is 3D produced? Why is some 3D good and some not-so-good?  How can 3D enhance storytelling?  Even if you don’t work in motion picture production you’ll find this information useful. 3D movies as with 2D movies are based on illusion. For 2D we project 24 frames of still images a second to create the illusion of a moving image.  Adding the illusion of volume and distinct separation to the moving picture is just as simple. All that is needed is two images each representing a slightly different point of view of a scene. These two points of view represent our eyes. In real life we see the world from two points of view: the left eye and the right eye. The fact that we see the world from these two points of view allows us to catch a baseball park a car or estimate distance on a golf course. If we only saw the world from one point of view these tasks would be very difficult. If you want to test this out don't go outside and try to catch a baseball with one eye closed although it would prove our point. You can simply close one eye and try to grab an object on the edge of your desk (the farther away the better). Did you miss it on the first try? Did you knock something over on the way? This happens because you can't judge distance without two different perspectives on the world. Watching a 2D movie is like seeing the world with one eye closed. You can't judge distance; it all looks flat. A 3D movie puts both eyes to work allowing you to judge the distance of every object on screen. It's an amazing trick for two flat images that appear to be similar when viewed individually. Of course we did leave out the one thing that everyone wishes wasn't the case. For the 3D illusion to work with projected images you HAVE TO wear glasses. Since both perspectives are projected at the same time on the screen 3D glasses ensure that each eye only sees one perspective. Take solace though: the glasses of today don't look nearly as geeky as those of your parent’s era. In fact Ray Ban is getting into the business of creating designer 3D eyewear. Now that we understand how the illusion works we can delve into how 3D is made. We know we have to see the world from two different perspectives so one solution is to slap a second camera next to the original one. This is called stereoscopic cinematography or 3D capture. Blockbusters such as Hannah Montana 3D The Final Destination 3D and much of Avatar were created in this way. Unfortunately capturing 3D is not really as easy as slapping a second camera next to your original one. Just as a 2D camera presents a set of parameters that the cinematographer must master two cameras working in tandem present additional parameters that must be mastered to capture the illusion in the desired manner. Another approach to creating 3D is with computer generated animation or CG 3D. Films like Up Cloudy With a Chance of Meatballs and large portions of Avatar were created with this approach. Similar to 3D live capture the CG world has to be captured from two perspectives.  But in CG the cameras are much easier to master. In fact the CG world can be rendered with more than one pair of CG cameras allowing the director to choose the most pleasing and comfortable effect. After all 3D doesn't re-create the real world. It's an illusion that requires tweaking to create a comfortable and compelling experience. The third approach to creating 3D is called 2D to 3D conversion. Films such as G-Force and Alice and Wonderland incorporate this method. Live action is captured from a single camera and through a visual effects type process a second view is generated to create the stereoscopic effect. A simple way to conceptualize this is to think of the left perspective as the original capture and the right as the newly created perspective. However the stereo perspective is created there are simple rules that govern how our eyes and brain interpret it.  If we shift the right image slightly to the left the viewer’s eyes will slightly cross. If you were to draw a line from each eye to where it’s looking on the screen those lines would cross somewhere in front of the screen. Where those lines cross is where the viewer perceives the image to be. If the right image is shifted to the right the lines from the viewer’s eyes would cross somewhere beyond the screen creating the effect that the image is beyond the screen plane. In 2D to 3D conversion this concept is applied for every object that requires distinct separation from the background. Creating shape is a little more complicated but it relies on this same concept. Both 3D conversion and CG 3D are similar in that every element within a scene can be tweaked and adjusted to get the desired effect. No matter how the 3D effect is created it is important to consider a number of things that will affect the filmmaker’s depth decisions. First and foremost what is the artistic goal? Is the purpose of the 3D to immerse the audience in the story? Is it to take them on a ride? This basic question will color many of the filmmaker’s depth choices. If the goal is to immerse the audience then ensuring audience viewing comfort proper scale and proper depth continuity is essential (more on these later). If the goal is to take the audience on a ride such as with a horror flick then how to get the most impact out of the depth is paramount. This may mean 3D gags that break the fourth wall superfluous angles and sequences or depth that’s not natural but makes for a fun ride. Editing pace is another important aspect to consider when making depth decisions. It takes much longer for the audience to take in an entire 3D shot than for a 2D shot. This is because the audience has to look at different points in space to actually see the entire shot. Often the faster the edit the less depth can be comfortably handled by the audience. This can be mitigated by ensuring very tight depth continuity or by reducing the overall depth of the shot to a comfortable viewing range. It does not hurt to let a shot linger a beat longer in a 3D movie. In many ways a 3D shot gets a lot more mileage than a 2D shot. Consider a scene that traditionally has a lot of cuts such as a car chase scene. Instead of relying on the editing pace to give the scene intensity depth can help create this feeling. Imagine a point-of-view shot from within a car that is careening down a sidewalk. It runs through newsstands and tables and people barely dive out of the way. That would be pretty intense especially if the shot lingers long enough for the audience to see what the car is driving through. The kinetic nature of a shot is also important to consider. A general rule is that the more kinetic or shaky the shot the less 3D is important or even wanted. Highly kinetic shots in particular can be uncomfortable to view in 3D if they have a lot of depth. Framing is another aspect to consider when making depth choices. Generally if a filmmaker wants something to come out of the screen it should appear in the center of the frame. If the object breaks left right or to a lesser degree the top of the frame it will limit the “off-screen effect.” This can be mitigated somewhat with a technique called a “floating window.” But this technique is limited as it can create the effect of looking down a tunnel if used to too great a degree. The framing approach described doesn’t just apply to gag shots. Even over-the-shoulder shots can benefit when an alternate framing is applied. Does the foreground character (the one whose shoulder we’re looking over) need to break the edge of the frame? If not then don’t cut them off at the edge of the frame else you will need to use a floating window. Another important depth decision pertains to realistic scale versus a high 3D effect. Often these goals oppose each other especially in the case of large objects contained within the frame (such as cityscapes or giant monsters). Creating a proper sense of scale is something that the visual effects industry often has to battle. Usually the question is how do you make a miniature of Mount Rushmore look as big as the real thing? But in 3D the real Mount Rushmore can be made to look like a miniature simply by giving it too much depth and shape. This is because in the real world Mount Rushmore will appear nearly “flat” due to size and distance. At great distance our eyes cannot perceive any difference between the left and the right view of an object. It’s ironic that many beautiful vistas need to play nearly flat in 3D. Such shots may sound not so compelling and forgettable but this does not have to be the case. Framing in closer objects will provide visual reference and emphasize the large scale of a distant object creating a good 3D effect from an essentially flat subject. In fact taking this approach with miniatures can help sell the fact that they are not miniatures. Beyond these considerations for making good 3D there are many ways to use 3D as a storytelling tool. 3D can provide a sense of immersion that allows the audience to suspend disbelief just a little bit better than with 2D. After all 3D is how we see the world. Artistically the filmmaker can quickly give the audience a sense of scale and geography. Also the filmmaker can create a more visceral experience for the audience. The audience can feel the confines of a tight space the crowd around them in a concert film or the height of a cliff. As your mind perceives the depth of the environment the unavoidable visceral experience comes as well. If the rocky bottom of a cliff appears hundreds of feet away the audience can't help but get sweaty palms. The experience and the potential of the medium are unparalleled. How objects are played in space can say volumes in the hands of a filmmaker. Staging characters in the frame has always been a way to inject meaning into a shot. The relative placement of characters in height can suggest dominance or the physical separation between a foreground and background character can mean they are emotionally separated. Taking the last example 3D can add even more meaning. The foreground character will naturally have more shape in terms of depth than the background character. Given the proper dialogue this staging can suggest that the foreground character is more emotional than the background character. If one imagines the same setup but now with a long lens so that the characters are the same size the background character will appear larger and more powerful. This is because the background character is positioned farther away from the audience in space. Distant objects like a mountain are larger in perception than the objects that are close to us. Since both characters projected the same size on screen our brain subliminally tells us that the background man must be bigger. This concept essentially capitalizes on the proper scale issue discussed earlier to create an artistic effect. Object placement can evoke a visceral response. A director of a horror movie may shoot a close up of a disgusting monster to make the audience cringe. In 3D that same close up can be brought closer to the audience to make them feel even more uncomfortable. In Dial M for Murder Hitchcock placed an object just out of the reach of the heroines grasp but out of the screen close to the audience’s grasp. The overall depth of a scene can also say much about a character’s view of the world. For example a mouse will view humans as huge as pine trees and a room of a house as spacious as a cathedral. POV shots can put the audience in the mouse’s shoes by creating or capturing the depth so humans are exaggerated to look as tall as pine trees. Of course this technique can be used to give the audience an insight to a character’s state of mind. If a character is embarrassed and just wants to disappear within their skin a filmmaker could use a high angle and exaggerated depth to make the character look small. Overall depth can also be used as a metaphor as it was in Pixar's Up. The sad scenes played flatter while the happy scenes played with a rich sense of depth. This is quite similar to how a filmmaker might use light to underline the same emotions. Many of the standard cinematographic approaches used today are very applicable to 3D and the purpose behind those choices should be examined. Low depth of field is used to focus the viewer’s attention on the subject. This can be underlined in 3D by minimizing the depth of the out of focus area making it less interesting to the eye. A monster pursuing someone may be shot with a long lens to make it appear as if the running person is going almost nowhere. The reverse of that may be shot with a wider lens to make it appear as if the road to escape extends out impossibly far. In 3D the first choice can be supported and emphasized by limiting the depth between the monster and the character or even falsely reducing the distance to make it appear as if the monster is almost on top of the character. The second choice can be supported by giving the road a great amount of depth making it appear incredibly long and escape impossible. The possibilities introduced by 3D are vast and exciting. As more and more films go 3D filmmakers will need to understand more than just the basics of how 3D is made but how to create good 3D while avoiding bad 3D and the artistic storytelling choices that 3D offers.  As a new technology 3D will continue to present creative challenges as 3D moves into the home.  3D is no longer just for nerds but a tool for storytelling that will bring enjoyment to audiences for years to come.