# Tag Info

4

If you are asking about online learning using OpenCV's Cascade Classifier, I'm pretty sure it is not possible. The reason is that Cascade Classifier consists of number of simpler classifiers which are decision trees in case of OpenCV's face detection system. I don't know exactly, what decision tree algorithm is used, but as far as I know there are no ...

3

Object detection is relatively a heavy task as you've notice. Detecting the object (in your case human face) in every and each frame would be cumbersome and computationally immense. Therefore, you need to employ an object tracking technique. There are various tracking algorithms, of which, KLT and mean-shift are the two popular ones. KLT works based on ...

3

We know that the problem is solved because our cell phones and digital cameras can do it. You might be able to find out what algorithm they use by trawling the patent databases. Computer vision libraries like OpenCV and SimpleCV offer face detection as a standard feature. See for example Near realtime face detection on the iPhone w/ OpenCV port. If you want ...

2

Here is one way to set up the project: Rather than analysing the difference between two pictures, try analysing the pictures themselves. For example by trying to identify charactaristics of people that make them unique. Once you do this once for the base image you can accept a login if the result of the new pic is sufficiently close to the baseline. After ...

2

Haven't used them first hand but I'm pretty sure that [SIFT] "Scale Invariant Feature Transform" is scale invariant. And I know that they are used, with great success, in face recognition.

2

Sure, that is one way of accomplishing your task so long as you do blending/interpolation to smooth the pixels in between. See these slides (esp. the last slide) or this survey for some literature on image warping and registration. Thin-plate spline or "surface spline" problem One reasonable approach would be establish point correspondences $r_1, \ldots, ... 2 The step-to-step explanation in Eigenface seems quite clear to me. A covariance matrix is like an high-dimensional extension of the variance, which is computed by removing the average from your only sample. Yes, you remove the average face ($ \operatorname{AF}\$ from all images, but keep it preciously. Your cov(image1) definition seems weird to me, but ...

1

Somebody did it for you: Viola-Jones Face Detection for Matlab Viola Jones Object Detection and I guess you can find a few others. A nice description, in pseudo-code, can be found in An Analysis of the Viola-Jones Face Detection Algorithm, IPOL, 2014, which you can follow to code your own.

1

I think you are talking about multiscale object detection. It will rescale the images to different resolution and then use trained classifier over it.

1

The first approach assumes that you already have identified local features, special points on the face. This identification task is not always straightforward to perform: imagine a face with sunglasses. So it requires quite sophisticated preprocessing. The second one uses more global features, by produce a set of optimal "fake images" (eigenfaces) built ...

1

What lies at the heart of pattern recognition and pattern classification is the selection of the correct features that is used in decisions. And the most important properties of correct features are 1-ease of measurement , 2-power of discrimination. So not every feature is the same. And finding the ones which yields the best performance requires research, ...

1

The Sum-area table (SAT) was introduced in computer graphics, and as far as I know, was popularized in computer vision and image processing under the name Integral image. Apart from a larger genericity of the SAT to objects of dimensions different to that of images, the concepts are the same to me.

1

You are correct, Haar features, as well as LBP and HOG are not rotation invariant. I have experimented with vision.CascadeObjectDetector in MATLAB's Computer Vision System Toolbox, and found that the face detector model it comes with can tolerate about 15 degrees of in-plane rotation. So if you want to handle in-plane rotation, the easiest thing is to ...

1

I know it is not matlab, but this post about training a 'banana' detector is really great and can give you some tips.

1

If you want to train the classifiers with your own database, you will only need 'trainCascadeObjectDetector' function and feed your images into the proper arguments (Positive, negative images). The output classifier will be in your 'outputXMLFilename' as in traincascadeobjectdetector trainingImageLabeler is helpful function for classifying positive/...

1

Someone's written a php interface to OpenCV - https://github.com/mgdm/OpenCV-for-PHP, though I haven't tested it. As you've tagged face detection I'll also mention that someone has written a php interface specifically to the face detection functionality of openCV too; http://www.xarg.org/project/php-facedetect/ Alternatively you could try (and this would ...

1

There are a few of things you can try: Definitely move FaceDetect = vision.CascadeObjectDetector; outside of the loop. You only need to create the face detector object once. Re-creating it for every frame is definitely your performance bottleneck. vision.VideoFileReader returns a frame of class 'single' by default. If you change the output data type to '...

1

It is an essential component in Principal Component Analysis(PCA) that allows to reduce the number of dimensions by projecting less dimensions of this sort of data, thus narrowing it down faster to finding a pattern. The projection operation characterizes an individual face by a weighted sum of the Eigen faces features and so to recognize a particular face ...

1

I will try giving you some intuition. The SVD says each matrix can be decomposed into 3 operations - Rotation, Stretching (Scaling) and the another Rotation. What matters is which directions are scaled and how. Directions are vectors (Pointing some direction). The SVD has many uses in Linear Algebra. One its most known use is Low Rank Approximation of a ...

1

Have a look at this report by Marques (PDF). In particular, have a read through section 1.4. It describes several different aspects to the face detection problem. You might want to decide what scenario you're after. Fully unconstrained (uncontrolled environment) face detection is hard to do accurately. As a first approach, try just doing color-based ...

1

Check out Yang and Ramanan's work with flexible mixture of parts. It starts with a central part, but pairs of parts that are attached are detected jointly by estimating the probability that the image area being queried supports the co-occurence of the attached parts in the image. Error is presumed in the attachment and is minimized using a spring-like ...

1

If you can transform your problem into the following "given a set of vectors find the N ones with the smallest euclidian distance to the query", there are well-known methods (Ball-trees, Locality-sensitive hashing...) for solving it on large datasets. With this approach, PCA / LDA is thus just a pre-processing step for transforming the set of features ...

1

FYI... STASM uses OpenCV's face detector (albeit very slightly modified for better image-boundary detections). If you really want a better face detector then I recommend hunting down detectors from the FDDB dataset challenge (most probably proprietary) or create your own custom detector to fit your needs. If you still plan on going the route of verification ...

1

I'd suggest Viola-Jones Face detector. Easy to implement and efficient in computation time. I did some face recognition project circa 2009 and this method was the most computation efficient method back then. Original Paper: research.microsoft.com/~viola/Pubs/Detect/violaJones_IJCV.pdf Basic Knowledge: http://en.wikipedia.org/wiki/Viola%E2%80%...

Only top voted, non community-wiki answers of a minimum length are eligible