The term "depth video" can be understood in a couple of ways, depending on whether you are referring to the visual illusion of depth in traditional video or a technical data stream that explicitly captures depth information.
Depth in Traditional Video (Creating the Illusion)
In the context of filmmaking and video production, "depth" refers to techniques used to make a flat, two-dimensional image appear three-dimensional. This isn't capturing literal distance data but rather manipulating visual elements to create perspective and layering.
According to the provided reference: "Depth in film refers to the ways you add layers to your shots to give the illusion of a three-dimensional image. Depth gives your audience the sense that they can see through the screen and into the world you created, as opposed to looking at a flat, two-dimensional image."
Creating this illusion in video involves various cinematic techniques, such as:
- Using foreground, middle ground, and background: Placing objects or subjects at different distances within the frame.
- Depth of field: Using focus to isolate a subject while blurring elements closer or further away.
- Lighting and shadows: Using light to create shape, form, and separation between elements.
- Atmospheric perspective: Including elements like haze or fog that naturally reduce contrast and detail in the distance.
- Movement: Using camera or subject movement to reveal parallax and spatial relationships.
- Blocking: Arranging actors or objects in three-dimensional space within the scene.
"Depth video" in this sense is simply standard video or film that effectively employs these visual strategies to make the image feel spacious and dimensional to the viewer.
Depth Video (Technical Data)
Separately, in technical fields like computer vision, 3D scanning, and augmented/virtual reality, "depth video" typically refers to a sequence of images where each pixel's value represents the distance from the camera or sensor to the surface of an object in the scene. This is often captured by specialized sensors, such as:
- Stereo cameras: Using two cameras slightly offset to calculate depth based on disparity.
- Structured light sensors: Projecting a known pattern onto a scene and analyzing its deformation to calculate depth.
- Time-of-Flight (ToF) sensors: Measuring the time it takes for a light signal to travel to an object and back.
A technical depth video stream is not a visual image in the traditional sense (like RGB color video) but rather a data stream where pixel intensity often corresponds to distance. This type of data is used for applications like:
- Creating 3D models of environments or objects.
- Gesture recognition and body tracking.
- Autonomous navigation for robots or vehicles.
- Enabling effects like background blur or object segmentation in real-time.
In summary, while traditional video uses cinematic techniques to create the illusion of depth as described in the provided reference, "depth video" in a technical context refers to a data format that explicitly captures and stores distance information for every point in a scene over time.
