opalmirror's profile - activity

There are two factors to consider here depending on the constraints of your application. the farthest distance for whi

You might try googling for 'opencv stereo depth to point cloud example' and reading thru example code to look for incons

The output of the stereo disparity computation is a 2D Mat of 16-bit signed values corresponding to the first input arra

I have a partial answer: 1) You may have two different applications of stereo matching here. You may need to filter the

I suppose you have noted a lack of depth solution in low-contrast areas of your object. Some ideas for adding unique reg

I suppose you have noted a lack of depth solution in low-contrast areas of your object. Some ideas for adding unique reg

I suppose you have noted a lack of solution in low-contrast areas of your object. Some ideas for adding unique registrat

Hollows are areas where the disparity was not able to be determined. It can be due to lack of distinguishable physical f

From Camera Calibration and 3D Reconstruction, in the Detailed Description section, you'll see that calibration fills in

Note that compute returns a 2d array with 16-bit fixed point disparity values, with the least significant 4 bits the fra

Note that compute returns a 2d array with 16-bit fixed point disparity values, with the least significant 4 bits the fra

The SGBM API can provide you a table relating disparity values to depths (presuming your cameras are calibrated correctl

If it's the correct answer, please check the checkmark so it can be closed as solved! :-) Thank you.

See edit to answer for details.

Generally, you can rotate a portrait image into landscape or vice versa (remap pixels) prior to feeding it to the calibr

To the point, I'd merely exchange fx and fy, for pi/2 and 3pi/2 rotations, and leave them as is for 0 and pi rotations.

To the point, I'd merely exchange fx and fy, cx and cy for pi/2 and 3pi/2 rotations, and leave them as is for 0 and pi r

I have a question: does the camera switch between portrait mode and landscape on its own, and then you feed the resultin

Generally, you can rotate a portrait image into landscape or vice versa (remap pixels) prior to feeding it to the calibr

You may want to compare output at each step in the chain - capture of initial images, result of disparity, result of WLS

I will go ahead and mark it correct for you then. Thanks for contributing a good question, discussion, and answering it

Oh, if you are satisfied with your own answer, you can mark it correct too, more karma points. :)

Thanks for posting your answer. So, the SVM decides "this histogram looks like the histogram from other faces, and not l

Wrong result in calibrateHandEye function I am working in a robot application, in which I have a camera fixed to a robot

Prostoi, I am sorry, this is an application area (not an opencv operational question). It's also not an area where I hav

Prostoi, I am sorry, this is an application area (not an opencv operational question). It's also not an area where I hav

You could try making a normalized histogram of depths over the whole image, or if you can somehow identify a face region

You could try making a normalized histogram of depths over the whole image, or if you can somehow identify a face region

See Python API for cv2.findChessboardCorners using google. You'll get to OpenCV: Camera Calibration and Reconstruction.

@JackInaBox thank you for the correct answer vote! Your ideas sound worth trying. I haven't used outlines to solve dis

@JackInaBox this is a good discussion. If your real subject is just the scene as pictured, you will have errors. - Perh

@JackInaBox this is a good discussion. If your real subject is just the scene as pictured, you will have errors. - Perh

The StereoSGBM algorithm (and to a lesser degree, StereoBM) is optimized to find common visual elements of locally flat

Experimentation is ALWAYS worth doing. There is little alternative to learning by doing, because this is a complex field

This should be easy for you to determine empirically (by trying it). From a mathematical perspective, what are your sett

You are seeing calculated depths, which, frame to frame, are calculated as very different disparities from your stereo c

You are seeing calculated depths, which, frame to frame, are calculated as very different disparities from your stereo c

I think you'll find imread did not read an image in, did not print any errors, and so an image with width and height of

Are these the left and right images which have been undistorted and rectified? Because that is what you need as input to

Calibration: Accurately calibrate your cameras (intrinsic and stereo extrinsics). Minimize noise: Stereo disparity algo

Calibration: Accurately calibrate your cameras (intrinsic and stereo extrinsics). Minimize noise: Stereo disparity algo

Calibration: Accurately calibrate your cameras (intrinsic and stereo extrinsics). Minimize noise: Stereo disparity algo

The stretching along the Y axis makes me think your calibration might have misdetected values stored in a K matrix (fx,

Does img contain data in a 3 channel color format? If not, you'll need to convert the color format Does img_counter hav