Stereoscopic vision

[SeanKetring]

Introduction

Two cameras, placed at a known distance from each other, could be used to figure out the distance to an object by comparing the apparent position between them.

This would work by checking the shift in apparent position.

Notes

• Start by figuring out the formula. Don't bother with code, just work out the trigonometry.
  • Hint: This is called parallax, and Astronomers use a similar method to estimate distance to stars by checking its position in the sky, waiting six months, and then seeing how far it shifted. Look there if you're having trouble with the math.

    Requirements

• A formula for estimating distance to an object from the parallax shift.
• An accurate estimation of distance to an arbitrary object.

  Resources

• Disparity explanation

[AhmedSamara]

(Issue updated, please view it on github)

[AhmedSamara]

@nv4

[ghost]

Going to be joining in on this

[TabaQ]

Same here

[Tushar1711]

Same here

[imaniaminah]

Same here

On Thursday, September 17, 2015, Tushar1711 notifications@github.com wrote:

Same here

— Reply to this email directly or view it on GitHub https://github.com/IEEERobotics/bot/issues/395#issuecomment-141261583.

Imani William BA Mathematics | BS Electrical Engineering Meredith College and NC State University | 2018 iawilliam@email.meredith.edu Mobile: (919) 428-0211

[fliesnow]

After some cursory investigation, we found that opencv actually has a stereoscopic depth mapper included in its software. https://opencv-python-tutroals.readthedocs.org/en/latest/py_tutorials/py_calib3d/py_depthmap/py_depthmap.html#py-depthmap

[imaniaminah]

http://www.intorobotics.com/fundamental-guide-for-stereo-vision-cameras-in-robotics-tutorials-and-resources/