Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>You can find a section on setting thresholds in the documentation on <code>cv::FaceRecognizer</code> at:</p> <ul> <li><a href="http://docs.opencv.org/trunk/modules/contrib/doc/facerec/facerec_api.html#setting-the-thresholds" rel="nofollow">http://docs.opencv.org/trunk/modules/contrib/doc/facerec/facerec_api.html#setting-the-thresholds</a></li> </ul> <p>It works just the same for the OpenCV Python Wrapper, which you can easily see when calling <code>help(cv2.createFaceRecognizer)</code> in Python:</p> <pre class="lang-none prettyprint-override"><code>Help on built-in function createEigenFaceRecognizer in module cv2: createEigenFaceRecognizer(...) createEigenFaceRecognizer([, num_components[, threshold]]) -&gt; retval </code></pre> <p>So in the code you would create the model with a threshold, I'll set it to <code>100.0</code>. Anything below this will yield <code>-1</code> in the prediction, which means this face is <code>unknown</code>:</p> <pre><code># Create the Eigenfaces model. We are going to use the default # parameters for this simple example, please read the documentation # for thresholding: model = cv2.createEigenFaceRecognizer(threshold=100.0) </code></pre> <p>As shown in the demo, you can get the prediction and associated confidence (which is the distance to the nearest neighbor in the your training dataset) with:</p> <pre><code>[predicted_label, predicted_confidence] = model.predict(image) </code></pre> <p>So if you train your model on the subjects <code>A</code>, <code>B</code>, <code>C</code> <strong>AND</strong> you are using a threshold, then a prediction for <code>D</code> should yield <code>-1</code>, while <code>A</code>, <code>B</code> or <code>C</code> should be recognized. Given the fact, that you are using a threshold. </p> <p>As for adding new faces iteratively without re-estimating the whole model. This is not possible for the Eigenfaces or Fisherfaces method. You always have to call <code>FaceRecognizer::train</code> for these two algorithms to learn the model. The Local Binary Patterns Histograms (LBPH) model, which you can create with <code>cv2.createLBPHFaceRecognizer</code>, supports updating a model without recalculating the other training samples. See the API Documentation on this:</p> <ul> <li><a href="http://docs.opencv.org/trunk/modules/contrib/doc/facerec/facerec_api.html#facerecognizer-update" rel="nofollow">http://docs.opencv.org/trunk/modules/contrib/doc/facerec/facerec_api.html#facerecognizer-update</a></li> </ul>
    singulars
    1. This table or related slice is empty.
    1. POeigenfaces with openCV
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USbytefish
    1. USbytefish
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POeigenfaces with openCV
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTAcceptedByOriginator
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COsorry, with "rerecognize persons that are not in the database" I meant that you are able to rerecognize an unknown face again after seeing it without retraining the model.
      singulars
      1. PO
      1. USnomoral
    2. COTo the people starting with opencv, `help(cv2.createFaceRecognizer)` will result in an error because there is no module `createFaceRecognizer`, took me a while before I gave up searching the source.
      singulars
      1. PO
      1. USIsaias
    3. COThis is not correct. You can find a full source code example on how to use the OpenCV Python Bindings and Eigenface Model at: https://github.com/Itseez/opencv/blob/master/samples/python2/facerec_demo.py. There is no method such as ``cv2.createFaceRecognizer``, this is the reason you get an error. Have a look into the API documentation linked in the above answer, to see which functions are available.
      singulars
      1. PO
      1. USbytefish
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload