https://github.com/edenton/drnet

https://github.com/edenton/drnet

Last synced: 9 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/edenton/drnet
• Owner: edenton
• Created: 2017-05-30T22:29:19.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2018-02-15T22:40:19.000Z (over 6 years ago)
• Last Synced: 2024-08-02T15:22:03.472Z (3 months ago)
• Language: Lua
• Size: 80.1 KB
• Stars: 105
• Watchers: 6
• Forks: 14
• Open Issues: 4
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# DrNet 

### [[paper]](https://arxiv.org/abs/1705.10915) [[project page]](https://sites.google.com/view/drnet-paper//)   

Torch implementation for Unsupervised Learning of Disentangled Representations from Video.

# Training 

To train the base model run:

```

th train_drnet.lua 

```

or the model with skip connections between content encoder and decoder:

```

th train_drnet_skip.lua 

```

To train an LSTM on the pose vectors run:

```

th train_lstm.lua --modelPath /path/to/model/

```

##  Training on KTH

First download the KTH action recognition dataset by running:

```

sh datasets/download_kth.sh /my/kth/data/path/

```

where /my/kth/data/path/ is the directory the data will be downloaded into. Next, convert the downloaded .avi files into .png's for the data loader. To do this you'll want [ffmpeg](https://ffmpeg.org/) installed. Then run:

```

th datasets/convert_kth.lua --dataRoot /my/kth/data/path/ --imageSize 128

```

The ```--imageSize``` flag specifiec the image resolution. Experimental results in the paper used 128x128, but you can also train a model on 64x64 and it will train much faster.

Now you're ready to train the DrNet model by running:

```

th train_drnet_skip.lua --dataRoot /my/kth/data/path/ --imageSize 128 --nThreads 2

``` 

Setting ```--nThreads``` utilizes multithreaded data loading and will speed up training significantly.