Begining Stages of Machine Learning

This Week We Made a tool for segmenting training data Experimented with using an SVM for segmentation Worked on stuff for easily segmenting/labelling planes in PCL + distinguishing things “on top” of the plane from the plane itself The training tool: The plot on the right is HSV->XYZ. You draw a box around the desired color groups, and use the resulting segmentation. This is less annoying than manually drawing on images. A SVM was trained on the segmentation data and it ran in ~ne
Continue Reading →

HOG Descriptor Implementation

HOG Descriptors can be uesd to identify an object with a sliding window. It is very slow, so the goal is to train an SVM on the data to speed up the process. This method is much more reliable then the color segmentation we had previously relied on. Below is a snapshot of the selection window in action:
Continue Reading →

March 16 Testing Day Notes

  This Week We Got one competition task working using point cloud data, in the pool Successfully bumped the buoy Proved that the cameras work pretty well in dark conditions Ralph got all of the bags on the sub onto the third computer in MIL, “WALL-E” Forrest added a bunch of features to TxROS at our request, what a guy!   Shoutouts To Ralph’s friend Ariel, for dutifully documenting our pool-day shenanigans, and somehow fishing a stray competition element out of the po
Continue Reading →