连续手语翻译（S2T模型部署）

作为连续手语翻译基准实验模型

Neural Sign Language Translation

Camgöz, Necati Cihan, Hadfield S , Koller O , et al. Neural Sign Language Translation[C]// IEEE Conference on Computer Vision & Pattern Recognition. IEEE, 2018.

源码地址(S2T模型)：https://github.com/neccam/nslt

1.下载：

git clone https://github.com/neccam/nslt.git

目录结构： 

nslt
├── BaseModel
│   ├── bvlc_alexnet.npy
│   └── download_alexnet_weights.sh
├── BestModel
│   ├── download_model.sh
│   ├── translate.ckpt-102000.data-00000-of-00001
│   ├── translate.ckpt-102000.index
│   └── translate.ckpt-102000.meta
├── Data
│   ├── phoenix2014T.dev.de
│   ├── phoenix2014T.dev.gloss
│   ├── phoenix2014T.dev.sign
│   ├── phoenix2014T.test.de
│   ├── phoenix2014T.test.gloss
│   ├