Overlapping object segmentation using CapsNet

Member	Email	Github
Zhiao Zhou	[email protected]	https://github.com/zhiaozhou
Jianghao Zhu	[email protected]	https://github.com/jz2575
Chenyu Xu	[email protected]	https://github.com/cx463

CapsNet/Capsule

Please refer to Geoffrey E Hinton's original paper Dynamic Routing Between Capsules

Architecture

Demo

This project also contains a demo website that classifies any overlapping digit image you upload

Paper

You can find the pdf file of our final paper here

Reference

• M. Abadi, A. Agarwal, and et al. Tensorflow: Large-scale machine learning on heterogeneous distributed systems. 2016.
• J. Ba, V. Mnih, and K. Kavukcuoglu. Multiple object recognition with visual attention. 2014.
• J. Chang and Y. Chen. Batch-normalized maxout network in network. arXiv, 2015.
• J. Goodfellow, Y. Bulatov, and et al. Multi-digit number recognition from street view imagery using deep convolutional neural networks. 2013.
• laodar. a tensorflow implementation for capsnet.
• A. Prochazka and M. Yadollahi. Image segmentation for object detection. 2011.
• S. Sabour, N. Frosst, and G. Hinton. Dynamic routing between capsules. 2017.
• L. N. Smith. Cyclical learning rates for training neural networks. 2017.