/quadric_lod

LOD algorithm for urban game scenes based on quadric error metrics

Primary LanguagePython

quadric_lod

Generates LOD mesh based on quadric error metric with heuristics. See the report document for details. preview

Usage

python decimate.py [-h] [-a] [-b BOUNDARY] [-t THRESHOLD] [-o OUTPUT] [-p] [-s SCAN] input n

Positional arguments

input Path to input mesh file, the OBJ file should be trianglulated and does not contain edges, normals, or texture coordinates

n Number of decimation iteration

Optional arguments:

-h, --help show help message and exit

-a, --area Turn off surface area heuristic

-b BOUNDARY, --boundary BOUNDARY Boundary constraint factor, use 0 to turn off

-t THRESHOLD, --threshold THRESHOLD Pair decimation threshold relative to mean extent of the mesh

-o OUTPUT, --output OUTPUT Path to output file

-p, --plot Plots loss history

-s SCAN, --scan SCAN Scan the entire loss curve, save model every [-s] steps after [n] steps

Requirements

numpy (for linear algebra)

matplotlib (for plotting loss curve)

tqdm (for progress bar)