Object Detection Resources.md

Object Detection Resources

There is a file in this directory titled ‘train base’. It’s available to help speed up the process of setting up a trainer. Just modify the values to suit your needs. Copy, Paste, and Go.

Image Datasets

• http://www.vision.ee.ethz.ch/datasets/index.en.html
• http://www.cvpapers.com/datasets.html
• http://www.cs.cmu.edu/~cil/v-images.html
  • Negative Images:
    • svn co http://tutorial-haartraining.googlecode.com/svn/trunk/data/negatives/
      • ~3000 images
    • https://github.com/openalpr/train-detector/tree/master/raw-neg
      • ~3000 images
    • download zip
    • http://www.csee.wvu.edu/~xinl/database.html
    • use wget -r -A jpeg,jpg,bmp,gif,png <link> to pull all the images from a site (may come in handy later)

OpenCV object recognition/training tutorials

Naotoshi Seo’s Tutorial

• http://note.sonots.com/SciSoftware/haartraining.html#w0a08ab4

Coding Robin Haar Training Tutorial(Personally,my favorite)

• http://coding-robin.de/2013/07/22/train-your-own-opencv-haar-classifier.html

Object Training FAQ

• http://www.computer-vision-software.com/blog/2009/11/faq-opencv-haartraining/

OpenCV Docs on TrainCascade

• http://docs.opencv.org/doc/user_guide/ug_traincascade.html

In-Depth Explanation of Some Parameters

• http://answers.opencv.org/question/7141/about-traincascade-paremeters-samples-and-other/

In-depth answers to how to train properly

• http://stackoverflow.com/questions/16058080/how-to-train-cascade-properly