7 comments

[ 2.8 ms ] story [ 25.4 ms ] thread
> Cubr is a project I completed in three weeks at the end of my introductory computer science class at CMU

This guy wrote a 7,900 line Python tool for his intro CS class. Call me impressed.

Functionality is a better measure of productivity than lines of code. The functionality is impressive. The code appears to be written in a Java style, which increases the line count versus a more Pythonic style.
There's a showoff at the end of every semester for the students in the class to show off their final projects. It's amazing to see people who, before coming to CMU, had never programmer in their life, program things like A.I. assistants, fully-featured games, and incredible vision-based tools. Probably one of the most (in)famous CS course at CMU depending on who you ask.
While reading through this I was thinking: this sounds like a 112 term project where a guy went ham, read back at the beginning and it was
> I gathered a fairly large set of data on the HSV and RGB values of each colored sticker in different lighting situations.

Wouldn't LAB have been a much better and easier choice?

Looks like a post from 2013, it should have "(2013)" appended to its HN submission title.
Can someome explain why him understanding Hidden Markov Models would've helped with the color identification process?