Skip to content
code of Temporal Unet
Branch: master
Clone or download
Latest commit 3a0fa2e Apr 29, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data first commit Apr 9, 2019
figs upload figs Apr 18, 2019
unet upload 1d unet Apr 9, 2019
weights add weights readme.txt Apr 18, 2019
README.md Update README.md Apr 29, 2019
test.py add test.py Apr 18, 2019
train.py add train.py Apr 10, 2019

README.md

Temporal Unet: Sample Level Action Recognition using WiFi

Data is here, pre-trained model on action detection is here, and pre-trained model on action classification is here.

Data description

train_data.mat

  1. train_data_amp: CSI amplitude of training data: 1116x52x192, 1116 CSI series, each series has 52 carriers, and 192 samples
  2. train_label_instance: action labels, 1116x192, 1116 CSI series, each series and 192 samples that are labeld from 0 to 6 (background + 6 actions).
  3. train_label_mask: action labels, 1116x192, 1116 CSI series, each series has 192 samples that are binaryly labeled.
  4. train_label_time: start time and end time labels. 1116x2, 2 is for start and end.

test_data.mat

similar to the above

Network

Temporal operations including 1d convolutions, 1D maxpoolings, and 1D deconvolutions sweep along the time axis of CSI series. network

AP Curves

network

An example

network

If this helps in your research, please kindly cite,

@article{wang2019temporal,
  title={Temporal Unet: Sample Level Human Action Recognition using WiFi},
  author={Wang, Fei and Song, Yunpeng and Zhang, Jimuyang and Han, Jinsong and Huang, Dong},
  journal={arXiv preprint arXiv:1904.11953},
  year={2019}
}
You can’t perform that action at this time.