Wanted to build a way to make training data for a future hack involving Machine Learning Point Cloud data

What it does

Lets you take or upload photos, Processes the photos, Turns the photos into a pointcloud, clears up the noisy data, tries to fill in the gaps, lets you view it on the web

How we built it

With a lot of different tools and language...mainly with dubstep and redbull though

Challenges we ran into

So many different ways to write the construct and noise algorithms. Also getting linux to install everything needed...cmake mainly

Accomplishments that we're proud of

It works

What we learned

How to drop the bass

What's next for Spherical Point Interpolation (Fillr)

Plotting for World Domination

Share this project: