3D Local Convolutional Neural Networks for Gait Recognition
This repository is the official implementation of:
3D Local Convolutional Neural Networks for Gait Recognition
Zhen Huang, Dixiu Xue, Xu Shen, Xinmei Tian, Houqiang Li, Jianqiang Huang, Xian-Sheng Hua
In this work, we present a new building block for 3D CNNs with local information incorporated, termed as 3D local convolutional neural networks. Our local operations can be combined with any existing architectures. We demonstrate the superiority of local operations on the task of gait recognition where 3D local CNN consistently outperforms state-of-the-art models. We hope this work will shed light on more research on introducing simple but effective local operations as submodules of existing convolutional building blocks.
News!!!
According to the policy of our company, the code is here.