Skip to main content

Listed

, home
FREN

Stereo vision

In French:
Vision stéréoscopique

Definition

A method that computes the depth of a scene from two images taken by two offset cameras, like our two eyes. The same point appears at slightly different positions in the two images; this difference, the disparity, is greater the closer the point is. Knowing the distance between the cameras and their focal length, one can deduce the distance to the point.

Checked on against the references below. · Written with AI · checked against the sources Source: OpenCV, tutorial “Depth Map from Stereo Images”.

Concrete examples

UBTECH’s Walker S2 has a binocular stereo vision system built into its head, using RGB cameras only. The maker says it relies on deep learning-based stereo depth estimation algorithms.

What it changes for you

Luxonis’s documentation says passive stereo needs good texture in the scene, and gives blank surfaces such as a wall or floor as an example. Active stereo on Luxonis’s OAK Pro cameras largely alleviates this problem with an infrared dot projector. As for range, a larger baseline or a longer focal length gives a greater depth for the same disparity.

Additional material checked on .

Related terms

References

Back to the glossary