Login Register



Auto-login on future visits

Forgot your password?

All videos


Building outlier removal in a classifier

Outlier removal is an important initial step for almost any machine learning project. In this real-world example, we show how to clean our data set and how to include outlier removal into the final classifier.

VISION14 talk: Machine learning for next generation vision applications

How can machine learning help us to develop next generation applications? Recording of our presentation at Industrial Vision Days conference accompanying VISION14 Trade Fair in Stuttgart, Germany.

Object detection

This video shows how to detect objects in an image using machine learning tools. First we use a classifier to detect candidate regions, and then a powerful segmentation tool to identify each object.

Local image features

Before we can experiment with classifiers we need to find a good data representation for our images. We may want to extract features that highlight relevant characteristics for our application. Very often we need to learn from local texture information, as we can leverage the spatial connectivity of the pixels. In this tutorial we will focus on extracting local image features and illustrate why it is important.

Detectors explained

A detector is a classifier that focuses only on one class of interest. It can be very handy especially when we have lot of samples of one specific target class and do not know much about the other classes. In this tutorial you will learn how to construct a detector in one-class scenario or when non-targets are also present.

Identify typical and rare samples

When we submit a new sample to a classifier, we learn the class to which the sample is assigned. What if we would like to be sure of our classifier decision. Does the classifier have a strong evidence that the sample belongs to the class? This video shows how to identify if a sample is a typical or rare example of its class.

Choose the classifier with cross-validation

How to reliably estimate the performance of a classifier? The cross-validation is a good tool for it. It can also be used to simulate a realistic classifier operation.

Learn from local texture

Learning from local texture is a must in several applications such as quality control or medical diagnostic. In this video we will learn how to build classifiers based on color and texture in a rock sorting application.

Why meta-data matters?

What information is useful when designing your classifier? This video shows how meta-data helps you to improve not only the classifier performances, but also your understanding.

Draw your classifier by hand!

Sometimes it is easier to draw a classifier by hand then to train it. perClass offers you an interactive tool to draw classifiers directly in the scatter plot. You can make the decision boundary exactly as you wish!

Page 1 of 2 pages  1 2 > | Next Page