Posts from the “Machine Learning” Category

Training & Tracking

Suppose you have a video containing a ball in all frames to be tracked. What’s the standard process of the tracking? Two steps: training, and tracking.

Training

Sample videos -> features (rough): use a detector (without feature descriptor) to extract the features from sample videos.

Features (rough) -> feature descriptor: train a machine learning algorithm (e.g. svm) with the features (rough), after which the machine learning algorithm can generate a feature descriptor describing the features of the sample videos.

Tracking

Video -> features (relative accurate): set the feature descriptor to the detector, and use the detector (now with feature descriptor) to extract features for the video.

Fitting Curves and Surfaces to Data in MatLab

Use the curve fitting toolbox: read doc of linear and nonlinear regression for the app, functions and tutorials:

For surface fitting, all you need is raw data of the x, y, z axis. The rest can be handled by the curve fitting app or functions. The fitted model shows an equation and the value of the parameters based on the data.

Here’re the two steps to load and fit the data:

  • Fit the loaded data
    • cftool: type ‘cftool’ to open the curve fitting app
    • Select the x, y, z data just loaded
    • Select the math module to fit