Skip to content
John De Witt edited this page Apr 2, 2018 · 1 revision

Capturing data involves repetitive steps. slcapture exists to automate these steps so that the operator can stay sane.

The two important steps are:

  1. displaying a specific pattern

  2. taking a timed image of said specific pattern

Given 1) a list of patterns and 2) images taken by a camera viewing each pattern, it's possible to reconstruct various properties of a scene. Color and depth are two commonly sought values to reconstruct. After data is collected, it can be processed by slcrunch.

A single "scan" consists of a folder (e.g. scan_0001) with one or more images, along with a list.yaml file which describes the order.

scan_0001/

scan_0001/img_0.jpg

scan_0001/list.yaml