time

[xiayizhan2017]

hi, when i run the demo, Timings[s]: mesh time(encode inputs) time(eval points) time(marching cubes) time(refine) time(simplify) class name
n/a 8.02325 0.015601 1.29366 1.439129 4.992291 0.249395 mean 8.02325 0.015601 1.29366 1.439129 4.992291 0.249395 There is 'The inference time of our algorithm with simplification and refinement steps is about 3s / mesh' in the paper. My GPU is TITAN Xp, and I don't know why it takes so much time. Thanks in advance!

[LMescheder]

Hi @xiayizhan2017, this might be related to your hardware (number of CPUs, memory, etc.). You can speed up inference by deactivating the refinement steps in the config. In most cases, the results without refinement are already quite good and refinement only has to fix some remaining problems (e.g. the wings for the airplanes).

For references, these are my timings:

	mesh	time (encode inputs)	time (eval points)	time (marching cubes)	time (refine)	time (simplify)
class name
airplane,aeroplane,plane	2.476421	0.004356	0.185622	0.286012	1.973629	0.020001
bench	2.604014	0.004670	0.275009	0.290625	1.975990	0.046294
cabinet	2.701229	0.004617	0.361059	0.275236	1.958236	0.085722
car,auto,automobile,machine,motorcar	2.517779	0.004301	0.233938	0.258221	1.968386	0.042734
chair	2.748108	0.004604	0.403309	0.259985	1.972186	0.092507
display,video display	2.692285	0.004752	0.335654	0.289729	1.974845	0.072015
lamp	2.485771	0.004298	0.231635	0.284154	1.922944	0.041966
loudspeaker,speaker,speaker unit,loudspeaker system,speaker system	2.869564	0.004551	0.493983	0.284767	1.953432	0.113080
rifle	2.414753	0.004707	0.130775	0.296746	1.966976	0.009909
sofa,couch,lounge	2.607800	0.004229	0.304244	0.258011	1.964742	0.063347
table	2.776753	0.004727	0.427196	0.282940	1.961718	0.084139
telephone,phone,telephone set	2.518772	0.004449	0.214532	0.280802	1.970168	0.038571
vessel,watercraft	2.467528	0.004449	0.171889	0.291079	1.969049	0.023563

[xiayizhan2017]

Thanks you for your reply!