Random_stair_branch

ARCH 703 Virtual Engagement

One start point One start point Two start points
gif1 gif2 gif3

Introduction

The inspiration is from the image/pixel-based stairs generated by neural networks, which was trained based on a dataset of over 2,000 stairs pictures. This project tried to emulate the style and randomness of the generated stairs by coding to create a spatial reconstruction.

Features

  1. Multiple start points can to set up to allow more than one stair branch to grow in the space.
  2. Branch-out statement can allow a single stair branch to develop a new branch.

Method

Random walks: A random walk is known as a random process, which describes a path including a succession of random steps in the mathematical space.

Relative neighborhood graph: an undirected graph defined on a set of points by connecting two points by an edge whenever there does not exist a third point that is closer to both two points than they are to each other.

Requirement

Install Rhino 7

MAC OS:

Install Atom

Install rhino-python in Atom

Quick Start

MAC OS:

To generate stairs in Rhino space:

  1. Run the StartAtomEditorListener command from Rhino
  2. Open random_stairs_v3.py in Atom. Press control + option + key R to run the script

To make animations:

  1. Run the StartAtomEditorListener command from Rhino
  2. Set up a new folder to save output images
  3. In random_stairs_animation.py, modefy the path (the path of the new folder) for saving output images at line 427:
    folder = '/your/folder/'

Then, run the script. 4. Use other applications to generate gif or video.

Posters

poster1 poster2