2
$\begingroup$

I'm very new in this field so my question may be very stupid. I apologize for this at first.

I'm trying to use a proper camera to run a SLAM algorithm. As suggested by some people, stereo camera may be a good choice. When I searched online, I found terms "stereo camera", "depth camera", "3d camera", and "rgb-d camera", and I feel confused. My understanding is "rgb-d camera" is more than "depth camera" because it has rgb channels while simple "depth camera" only makes white-black images. However, I have no idea about the difference from "stereo camera" and "3d camera". I know some camera options such as ZED, BumbleBee, Kinect, Intel RealSense, to me they are all "stereo camera".....

Anyone can give me some instructions? Thanks a lot in advance!

$\endgroup$
4
$\begingroup$

3D/Depth-Camera:

The most inclusive terms, only means that the camera gives you 3d data.

Stereo-Camera:

One kind of 3D-Camera, consists (in most cases) of two gray-scale Cameras.

RGBD-Camera:

A sensor that gives you depth and color. In most cases, this refers to a Kinect-Style camera (or Primesense, Realsense), But a stereo-camera could also create colored pointclouds (if one camera is a rgb camera or if you have an additional camera).

You then also have the distinction of passive setups (like a standard stereo camera) or active cameras with some kind of projectors.

Active cameras can use a speckle projector to create a texture that is then captured by a camera (Kinect1, Primesense) or use flash and Time-based computations to measure the distance (Keyword is Time of Flight) (Kinect2, Picoflex).

$\endgroup$

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.