Vanishing-Point Method
Definition
Using the vertical vanishing point and two horizontal vanishing points to compute the camera's effective viewing geometry, then projecting the person's image position to a metric height independently of any specific reference object in the frame.
- Field
- CCTV-based height estimation
- Inputs
- One vertical and two horizontal vanishing points
- Advantage
- Does not require a known reference object in the frame
- Output
- Metric height computed from camera geometry alone
Common questions
How does this method avoid needing a reference object of known height?+
The three vanishing points fully define the camera's viewing geometry, including its effective height and tilt relative to the ground plane. Once that geometry is recovered, a person's image position can be projected directly to a real-world height without comparing them to a doorway or other calibration object.
What scene features does the method depend on being present and identifiable?+
It needs clearly visible sets of parallel lines in at least two horizontal directions plus vertical lines, typically from building edges, kerbs, or floor tiling. A scene without such structured lines, such as open ground with no built features, cannot supply the vanishing points the method requires.
Related terms
- Ground-Plane Calibration
- A physical survey of the scene after the event, measuring reference points on the floor relative to the camera position, allowing the...
- Perspective Distortion Error
- The apparent change in size of objects at different distances from the camera due to projective geometry. A person closer to the...
- Pixel Height Ratio
- The number of pixels occupied by a vertical extent (person, door, markings) in the image. Ratios between subject and reference pixel heights...
- Reference-Object Method
- Height estimation by measuring the pixel heights of both the subject and a reference object of known stature in the same frame,...
- Stature Estimation Uncertainty
- The combined error from pixel localisation, reference height knowledge, positional correction accuracy, and camera model assumptions, expressed as a confidence interval at...