search

shunsukesaito / SCANimate

This repository contains the code for the paper "SCANimate: Weakly Supervised Learning of Skinned Clothed Avatar Networks"
https://scanimate.is.tue.mpg.de/
Other
274 stars 31 forks source link

运行时间 #11

Closed HDYYZDN closed 2 years ago

HDYYZDN commented 2 years ago

想请问一下，您训练多少数据，花费了多久的时间?

JinlongYANG commented 2 years ago

我们在CAPE(https://cape.is.tue.mpg.de/) 的三维扫描数据集上做的训练，每个套装训练一个模型，每10帧选取一帧做为训练数据。其中00134和03375两个实验对象只用了trial1的数据。每个模型的训练时间在1天到2天之间，取决于每个套装的序列长短。 We train subject/outfit specific models on raw scans of each subject from CAPE dataset (https://cape.is.tue.mpg.de/). We sample the frames and use every 10th frame as training data. For subjects 00134 and 03375, we use scans from trial 1. Each model is trained for one to two days, depending on the number of frames for each outfit.