Our eyes perceive the world around us as a three-dimensional (3D) image. By 3D image, we mean that we see widths, heights and depths. In a TV show or a cinema movie, we also perceive the depth of these images although it is rendered on a two-dimensional (2D) screen (just width and height) !
How is it possible that we see 3D world on 2D surfaces like computer screen, cinema, or TV monitors ? In other word, how depths are interpreted to a screen ?
In fact, there are a lot of depth cues that we, as humans, sense the three dimensions of the physical world. Each of our eyes has a two-dimensional retina on which images of the three-dimensional world are sensed. But what about 3D movies ? Why do we perceive them as more three dimensional than other regular movies ?
In this post, I’m going to discuss what forms a 3D image ..
Depth is how far things are from us. Depth cues are information added to images to give us the illusion that these images are 3D. We may put depth cues into four major categories: Monoscopic, Stereoscopic, Motion and Physiological cues. Let’s discuss each category here:
Monoscopic Image Depth
Human beings typically have two eyes. Each eye perceive an image of the scene and it is the role of our brain to interpret the two images that come from each eye. The fact that we have to eye is what adds a depth (dimensionality) to the scene we see.
However, many depth cues don’t rely on having two eyes available and work with even one eye. Let’s preview some of them:
Interposition: it means that part of some object is hidden behind some other object. In the picture, we see part of the red rectangle is hidden behind the blue one. Our brain understands that the blue rectangle is in front of the red rectangle.
Shading: when light is blocked by objects, shadows give us clues as to where those objects are in a three-dimensional world.
Size: we see the same object in different sizes according to its distance from us. The far the object is, the smaller its size. We carry a memory of the object size and when it gets smaller, we perceive it as it is moving far away which gives us the illusion of depth. See this video and see how the rocket gets smaller when the land camera moves away or the rocket goes up.
Linear Perspective: have you seen the two lines of that a train goes on? Exactly! Linear perspective is about roads getting closer together. Thus, if there is something along the side of the road where is the road is narrower, you perceive that the object is farther back in the scene than other things in the road where it is wider.
Height in the visual field: when you look at the horizon, it looks higher in your visual field than other objects.
Atmospheric effects: if there is fog in the air, things in the distance will be less visible than things that are near you.
The above effects are the one easy perceives as 3D effects even with one eye. Remember that our brain receives two images of the same scene and interprets them to form the 3D scene.
Stereoscopic Image Depth
Pointing out to the previous question we asked in the start of this post; why do we perceive 3D movies as more three dimensional than other regular movies ? and here comes the answer. Stereoscopic Image Depth (or Stereopsis) is the depth cue that we get from seeing the same scene from two slightly different perspectives. Each eye sees a slightly different image than the other. Our brain analyzes these differences and uses them (with other depth cues) to figure out the relative locations of different elements of the scene.
Stereopsis is is much stronger for objects that are close to the observer and relatively non-existent for very distant objects.
In 3D cinema, what happens is that you wear a glasses that separates an image into two different ones. One frame for the right eye and one for the left one. These two images are slightly different. The glasses ensure that the image that is computed based on the perspective of the left eye is displayed to the left eye and the image that is computed based on the perspective of the right eye is displayed to the right eye. And that’s why it looks more 3D.
Look at some object in front of you! Move your head slightly to get different perspectives. This way provides you a strong cue of relative relative distance to objects. There is a parallax that leads closer objects appear to move faster than more distant objects in the view. Even without moving your head, motion cues also exist. For example, if you are looking at two roads that cross your visual field of vision, cars moving in the closer road appears faster that others moving on the farther one.
There are two primary physiologically depth cues based on the changes a body undergoes when observing objects at different distances.
Accommodation: when you look at an object at a specific distance, your eyes focus on that object. It requires a mascular change on objects at different distances. Give it a try now! This is exactly the same as the fact that you need to adjust the focus on a camera lens for objects at different distances.
Convergence: when you look at something very close, your two eyes are aimed more toward each other than when you are looking at objects farther away. This provides information to your brain regarding the distance of the object you are looking at.
Combining all (or some) of the above depth cues gives us the illusion that the scene is three dimensional. And this explains why we can perceive 3D scenes on 2D images.
I will be more than happy to receive your valuable comments 🙂
Source: Understanding Augmented Reality, Concepts and Application – Alan B. Craig