I am testing Aggregated Channel Features ACF [1] for mode than 6 months. Features are extracted from 10 channels, L LUV, U LUV, V LUV, 0° Grad, 30° Grad, 60° Grad, 90°Grad, 120° Grad, 150° Grad, and Mag Gradient. This 10 channels are relatively simple to extract mainly by sobel derivatives (cv::sobel). Yes, the implementation is mainly based on Opencv.

I am using this model param int modelRows = 88; int modelCols =modelRows/4;

Init features are randomly generated. This is funny part. In some cases program hold 5GB is Ram memory and can cause shutdown of my PC. I cant find why. :)

Now, I am trying different approach of weak feature selectors by AdaBoost, GentleBoost, WaldBoost .

This is only personal research. In our applications. Waldboost on Haar + LBP features + Kalman are fast and good enough for me.

I know that Opencv Contrib module has ICF ACF feature extractor and Detector learned by waldboost. After some testing i try to implement my own version and this is the first results.

Adaboost ACF learning details

  • I am using Adaboost 1000 weak classifier,
  • Ped dataset is my own. 10 000 positive samples. I am working on this dataset time to time for more than 2 years.
  • 8 000 Neg samples are generated randomly from my Travel pictures :)

Yes first results after 6 months.

Aggregate channel features first results

In this example is used Town Center Dataset only for demonstration. I can not find any condition of usage on this page. http://www.robots.ox.ac.uk Town Centre Dataset

[1] P.Dollar, R.Appel, S.Belongie and P.Perona. “Fast feature pyramids for object detection”. TPAMI, 2014. 1, 2, 7