BAE-NET: Branched Autoencoder for Shape Co-Segmentation
<BAE-NET>
BAE-NET: Branched autoencoder for shape co-segmentation目录
编者按
Learning Implicit Fields for Generative Shape Modeling (CVPR2019) 的续作,inside / outside indicator作为shape表征
Motivation
- 把形状的
co-segmentation看做表征学习问题 - 可以无监督、弱监督、
one-shot learning,只需要用几个exemplars,就可以在shape 分割任务上好过在分割shape上训练的SOTA - 无监督的 co-segmentation

overview
- 就是在Learning Implicit Fields for Generative Shape Modeling 的基础上,从原来的单个inside / outside indicator变成
k个inside / outside indicator (branched output, one neuron each) ,然后在最后max pooling 把几个neuroncompose在一起。 
- 让网络 “自动” 学出来一个个natural shape的neuron;没有强制保证



