You can place additional downloaded models in data/.
bunny.off is provided for your convenience.
Here are some suggestions:
Make sure to compile in release!
mkdir build; cd build
cmake -DCMAKE_BUILD_TYPE=Release ..
make -j 8
To display statistics about the results, you need to have pandas available.
cd python
./benchmark.py -c distance_to_mesh -o dist.json ../data
./benchmark.py -c ray_tracing -o ray.json ../data
Output statistics:
./stats.py dist.json
count mean std min 25% 50% 75% max
method
geogram 17.0 0.122634 0.090985 0.018138 0.057599 0.085547 0.213004 0.284186
morton_binary 17.0 0.179299 0.136215 0.039204 0.072267 0.117081 0.303362 0.457230
hilbert_binary 17.0 0.179900 0.137642 0.038078 0.071600 0.116555 0.316527 0.454705
hilbert 17.0 0.187877 0.146730 0.033257 0.072034 0.121204 0.321534 0.477837
ours_binary 17.0 0.210401 0.173534 0.020379 0.097633 0.133029 0.316605 0.591937
ours 17.0 0.222995 0.181206 0.028541 0.100343 0.140565 0.334233 0.594097
morton 17.0 0.240368 0.211510 0.022894 0.108532 0.164132 0.406410 0.784553
igl 17.0 0.293843 0.272498 0.032069 0.110822 0.154889 0.475230 0.914263
./stats.py ray.json
count mean std min 25% 50% 75% max
method
embree 17.0 0.247782 0.102379 0.084333 0.190652 0.219816 0.295780 0.423347
geogram 17.0 0.637124 0.332892 0.124273 0.369079 0.658865 0.826838 1.360295
ours 17.0 0.685104 0.386073 0.182225 0.357059 0.748395 0.880797 1.489862
morton 17.0 0.719434 0.356301 0.156949 0.424598 0.750109 0.963619 1.465881
hilbert 17.0 0.732410 0.377974 0.154870 0.428783 0.749345 0.965261 1.558961
ours_binary 17.0 0.965989 0.574602 0.239528 0.491594 0.995952 1.220360 2.237329
hilbert_binary 17.0 1.037296 0.547904 0.200241 0.575182 1.047844 1.337041 2.173295
morton_binary 17.0 1.052529 0.563368 0.203320 0.576216 1.057456 1.339751 2.178193
igl 17.0 3.528063 3.216025 0.501061 1.010425 3.157664 5.004837 13.671037