Dima Lisin

MathWorks

391 total contributions since 2013

I am an engineer at Mathworks working on the Computer Vision System Toolbox.

Contact

Dima Lisin's Badges

View details...

Submitted

Detect and Track Multiple Faces
Detect and track multiple faces from live video.

4 years ago | 16 downloads |

Answered
rectifyStereoImages is cutting off too much of my image
Set 'OutputView' to 'full', when you call rectifyStereoImages.

5 years ago | 0

Answered
Ambiguous Disparity Map and Inadequate 3D Scene Reconstruction
Hi Yildiray, There are several problems with your calibration: # The checkerboard is not really flat. You have to glue it ...

5 years ago | 1

| accepted

Answered
Finding multiple roots of a polynomial
roots().

5 years ago | 0

Answered
Camera Calibration for measuring planar object
Not sure what you are asking... You would need to write your own code to detect the object of interest. If it has a distinctive ...

5 years ago | 0

Answered
Robust background substraction in outdoor environment
Use |vision.ForegroundDetector| from the Computer Vision System Toolbox.

5 years ago | 0

Answered
how can i save the image and roi to train classifier
Try using the <http://www.mathworks.com/help/vision/ug/label-images-for-classification-model-training.html Training Image Labele...

5 years ago | 0

Answered
Error in the code
I am guessing that your image is M-by-N-by-3 (RGB). |detectSURFFeatures| only takes M-by-N grayscale images. Convert your images...

5 years ago | 0

Answered
Why does my videoplayer hang at the first frame while streaming live camera.
Sounds like the size of your frame has changed. Most likely it went from being M-by-N grayscale to M-by-N-by-3 RGB or vice versa...

5 years ago | 2

| accepted

Answered
How to connect Kinect 2 with MATLAB image acquisition tool?
Image Aquisition Toolbox now <http://www.mathworks.com/hardware-support/kinect-windows.html supports Kinect v2.>

5 years ago | 1

| accepted

Answered
3D reconstruction of points from three views - is there a 3 camera alternative to triangulate?
The |triangulate| function can only handle two views. However, the algorithm that it uses can be easily extended to multiple vie...

5 years ago | 2

| accepted

Answered
How can I calculate the actual distance in real world of an object from its picture that picked up by a camera?
With one camera, you may be able to compute the distance if you have a calibrated camera, and you know the actual size of the ob...

5 years ago | 0

5 years ago | 0

| accepted

Answered
How to use calibration parameters from KITTI?
The file |calib_cam_to_cam.txt| contains parameters for 3 cameras. * S_0x: is the image size. You do not really need it for ...

5 years ago | 0

| accepted

Answered
Problem with Computer Vision tool box
Hi Ali, Try the following: Go to "Preferences". Select "Image Processing Toolbox" on the left. Uncheck "Enable hardware op...

5 years ago | 2

| accepted

Answered
is there an algorithm of matching based on intensity information
Actually, you are not quite right. What you typically match are SURF descriptors, which are based on gradient orientation distri...

5 years ago | 0

Answered
Vision Toolbox Question: Is there a way to access the z-coordinate from the camera calibration process?
Hi Thomas, In the checkerboard's coordinate system the points on the checkerboard all have z=0. If you want the checkerboard ...

5 years ago | 1

| accepted

Answered
How to extract shape characteristics from an image?
|regionprops| function in the Image Processing Toolbox can compute various shape characteristics of regions, such as area, perim...

5 years ago | 0

Answered
How to save a point cloud object??
Hi Priya, This confusion is an unfortunate result of a name conflict. |reconstructScene| returns an M-by-N-by-3 array of xyz...

5 years ago | 1

| accepted

Answered
In two processed images, one shows 3 centroids and another shows 4 centroids. How to eliminate this extra centroids from centroids array.
Hi Naseeb, The |triangulate| function can return reprojection errors, which can tell you whether or not a match is correct.

5 years ago | 0

| accepted

Answered
Feature Based Panoramic Image Stitching
Hi Alex, |fullfile| is a function to create a path to your image folder. It simply concatenates the strings you give it with ...

5 years ago | 1

| accepted

Answered
Error "Subscript indices must either be real positive integers or logicals" on a matlab based example
This error means that |matchFeatures| didn't find any matches and |indexPairs| is empty. If it is expected that sometimes you do...

5 years ago | 0

| accepted

Answered
Foreground Detection Using GMM on Depth Images
The |step| method of |vision.ForegroundDetector| takes an image, which can come from anywhere. So you can just read your depth m...

5 years ago | 0

Answered
How can I calculate area of target object with Calibrated Camera
That would depend on the object. If it is circle, and you can measure its diameter, then you can calculate its area as pi*r^2. I...

5 years ago | 0

Answered
Hi, I doing a project that load a roadsigns.mat, which a exported from Image Lableler. Can someone help me to fix the error below ?
You should export the ROIs using the "Export" button. That will create a variable in the workspace that you can pass into trainC...

5 years ago | 1

| accepted

Answered
I calibrated my camera using Camera Calibration app,How do i apply it to my camera for further use
If you click the "Export" button on the app, it will export a |cameraParameters| object to your workspace. There is a variety of...

5 years ago | 0

| accepted

Answered
How to use undistortImage function ?
Hi Lise, You have three different questions here. Let's take them one at a time: # *Crash when using undistortImage.* Ca...

5 years ago | 0

Answered
A toolbox to recognize object/faces in a video
If you are still working on this, try the <http://www.mathworks.com/matlabcentral/fileexchange/47105-detect-and-track-multiple-f...

5 years ago | 0

| accepted

Answered
Intensity based background subtraction for effective object detection
Try using |vision.ForegroundDetector| object in the Computer Vision System Toolbox.

5 years ago | 0

Answered
How to map the disparity image values to the original image?
You would have "un-rectify" the image, and there is no easy way to do that because rectification is a non-linear transformation....

5 years ago | 0