I have tryed hough circle algo but it give not acceptable out put of detecting
eyes . Can we find way to get the eyes detect in image I having face front ...
Try this implementation from Oxford's Visual Geometry Group. Their pipeline consists of a Viola-Jones face detector at the front and then 9 face landmarks including inner and outer corners of both eyes, center/edge of nostrils, and mouth corners.
I've used this with some success in the past. If you want an exact eye corner, then some additional refinement may be necessary as this includes a model that tries to make a best fit for the entire face.
Related
I try to begin new game development that has rotating circle that has one or many holes as in the attached image , the problem is I need to use circle collides for the big circle., in addition I need to use small collides for the smaller holes to prevent and particles(smaller circles ) input the circle and to count no of collisions occurred
enter image description here
The circle collider itself does not have this functionality since it can only check collisions between perfect circles. (Behind the scenes all circle colliders just check the distance between points and these distances do not work with holes) The best solution I can think of would be to use a polygon collider and make a rough circle that you can then remove parts of (would require some advanced geometry but not impossible). I have tried simillar things in 3D and with the mesh collider, removing parts of it using a method known as CSG (Wikipedia) and something simillar should be possible in 2D however it is really, really hard and even experienced programmers struggle to come up with a system that can handle every scenario.
All hope is not lost though, depending on what you need the collision for, there might be other ways. If you just need to check if two objects intersect and don't want rigibodys and stuff to interact with the collider just checking the distance between the big circle and the object (too figure out if it's close enough) and then checking the distance between the object and the holes (too see if it's inside a hole on thus not colliding) is the best way to solve the issue.
I'm sure someone who have done more in this field than I have will be able to help you out but without more information of what you need it's impossible to say. It all depends on the circumstances!
I am using an API to analyze faces in Matlab, where I get for each picture a 3X3 rotation matrix of the face's orientation, telling which direction the head is pointing.
I am trying to normalize the image according to that matrix, so that it will be distorted to get the image of the face's plane. This is something like 'undoing' the projection of the face to the camera plane. For example, if the head is directed a little to the left, it will stretch the left side to (more or less) preserve the face's original proportions.
Tried using 'affine2d' and 'projective2d' with 'imwarp', but it didn't achieve that goal
Achieving your goal with simple tools like affine transformations seems impossible to me since a face is hardly a flat surface. An extreme example: Imagine the camera recording a profile view of someone's head. How are you going to reconstruct the missing half of the face?
There have been successful attempts to change the orientation of faces in images and real-time video, but the methods used are quite complex:
[We] propose a gaze correction method that needs just a
single webcam. We apply recent shape deformation techniques
to generate a 3D face model that matches the user’s face. We
then render a gaze-corrected version of this face model and
seamlessly insert it into the original image.
(Giger et al., https://graphics.ethz.ch/Downloads/Publications/Papers/2014/Gig14a/Gig14a.pdf)
I have a projector component and I need to find the angle that projected texture falls at to exclude the projecting on vertical faces.
My projector is under the mouse pointer and works ok when it is over an horizontal face:
I would like the projector to switch off on vertical faces to avoid this bad effect:
If possible, I would like to do it in the shader code to avoid the vertical projected image even if the cursor is located on the corners of an horizontal face and a part "goes out" on vertical face.
I found this solution in C#:
if (Physics.Raycast(MouseRay,out hitInfo)){
if(hitInfo.normal.y>0) {
// draw
} else {
// not draw
}
}
But only it works on curved surfaces and not, for example, on the face cubes.
How can I do this properly?
Normally they would use an image on a quad using TGA transparency, which rotates itself to the face that the middle of the object is aligned to, using ray to find the vertex and making it's absolute normal.
Other ways of doing it would be quite tricky, perhaps using decals... If you did it using a shader, it would take so much time... it's a case of problem solving not being ordered in order of importance for fast development. Technically you can project a volumetric texture on top of whatever object you are using... that way you can add your barred circle projected from a point in space towards the object, as a mathematical formula, it takes a while to do, check out volumetric textures, i have written some and in your case it needs the mouse pos sent to texture and maths to add transparent zone and red zone to texture. takes all day.
It's fine to have a flat circle that flips around when you change the pointer onto a different face, it will just look like a physical card and it's much easier to code, 10 minutes instead of many hours.
Maybe I'm asking this too soon in my research, but I'd better know if this is possible sooner than later.
Imagine I have the following square printed on a paper on top of a table:
The table is brown, so it does not match with any of the colors in the square. Is there a way for me, from a common iPhone camera (non-stereo view), to figure out the distance and angle from which Im looking at the square in the table?
In the end what I'm looking for is being able to draw a 3D square on top of this one using the camera image, but I'm not sure if I am going to be able to figure out the distance and position of the object in space using only a 2D image. Any hints are well appreciated.
Short answer: http://weblog.bocoup.com/javascript-augmented-reality
Big answer:
First posterize, Then vectorize, With the vectors in your power you may need to do some math tricks to define, based on the vectors position, the perspective and then the camera position.
Maybe this help:
www.pixastic.com/lib/docs/actions/posterize/
github.com/selead/cl-vectorizer
vectormagic.com/home
autotrace.sourceforge.net
www.scipy.org/PyLab
raphaeljs.com/
technabob.com/blog/2007/12/29/video-games-get-vectorized/
superuser.com/questions/88415/is-there-an-open-source-alternative-to-vector-magic
Oughta be possible. Scan the image for the red/blue/yellow pattern, then do edge detection to figure out how warped the squares are (they'll be parallelograms in anything but straight-on view). Distance would depend on the camera's zoom setting and scan resolution. But basically you'd count how many pixels are visible in each of the squares, run that past the camera's specs and you should be able to determine a rough distance.
I have two images. In one of the images, my eye is in the center position and in the other image, it is in the left. How do I find out whether my eye is in the left or the right?
I am using MATLAB. Are there any functions for this?
A simple solution is to try to detect the iris using circular Hough Transform.
You can find a lot materials out there. To name a few, these two fileexchange submissions:
Hough Transform for circle
detection
Circle Detection via Standard Hough
Transform
This sounds like Eye tracking implemented in MATLAB which is a fairly popular research topic.
If you want a more detailed answer, please answer the following questions:
Do you know the coordinates of your eye in the first image?
What kind of motion is there between the two images? Rotation/translation/scaling/...?
Do you want this to be real-time?
What is the resolution of the images?
Are there going to be more eyes in the image apart from yours?
If you are willing to select the eye in one image you can use template matching to find it in others (for example you can mark it in the first frame of a video and then find it in all other frames).
Look at the normxcor2 function in matlab:
http://www.nd.edu/~hpcc/solaris8_usr_local/src/matlab6.1/help/toolbox/images/normxcorr2.html
This technique is robust to constant illumination change, but will fail if the appearance of the eye changes significantly between the image you took the template from and the image you are searching in.
If you are going to search for the eye in a lot of frames (for example, eye tracking from a webcam) then you should look at stronger techniques such as the Kalman Filter or the Particle Filter (aka Condensation Filter in computer vision)
By using Color Distance Maps, the skin and non skin area can be differentiated and thus the non skin area contains the iris. From the iris, the whole eye could be detected. Hope it works.
You should also have a look at Eye Ball Detection in MATLAB , they have detected eyes first and then detected the EyeBall.