WSD and OpenCV

I've spent some time experimenting with automatic Word Sense Disambiguation/Distinction/Induction. I've learned a bit while working on it, though not what I intended. I didn't come up with any results that haven't been explored before. I did create a 10000x10000 image of word relations with Graphviz. Most recently I've been playing with OpenCV. OpenCV is an open source computer vision library. I'd like to teach it to recognize my family.