This repository holds the codes and models for the papers
SOS: Sequence-of-Sequences Model for 3D Object Recognition Tan Yu, Zhou Ren, Yuncheng Li, Enxu Yan, Ning Xu, Jianxiong Yin, Simon See, Junsong Yuan
These instructions will get you a copy of the project up and running on your local machine for training and testing purposes.
Pytorch
conda install pytorch torchvision -c pytorch
You need download ModelNet40 dataset
with orientation assumption, 6-view settings
wget http://www.cim.mcgill.ca/dscnn-data/ModelNet40_rendered_rgb.tar; tar -xvf ModelNet40_rendered_rgb.tar
without orientation assumption, 20-view settings
wget https://data.airc.aist.go.jp/kanezaki.asako/data/modelnet40v2png_ori4.tar; tar -xvf modelnet40v2png_ori4.tar
tar -xvf cache_model.tar
6-view ModelNet40
python mainalex.py -d modelnet40 -v 6
6-view ModelNet10
python mainalex.py -d modelnet10 -v 6
20-view ModelNet40
python mainalex.py -d modelnet40 -v 20
20-view ModelNet10
python mainalex.py -d modelnet10 -v 20
6-view ModelNet40
python testalex.py -d modelnet40 -v 6 --resume cache_models/alexnet_modelnet40_6view.tar
6-view ModelNet10
python testalex.py -d modelnet10 -v 6 --resume cache_models/alexnet_modelnet40_6view.tar
20-view ModelNet40
python testalex.py -d modelnet40 -v 20 --resume cache_models/alexnet_modelnet40_6view.tar
20-view ModelNet10
python testalex.py -d modelnet10 -v 20 --resume cache_models/alexnet_modelnet40_6view.tar