TTCF: Training Toolbox for Caffe

This is a BVLC Caffe fork that is intended for deployment multiple SSD-based detection models. It includes

• action detection and action recognition models for smart classroom use-case, see README_AD.md,
• person detection for smart classroom use-case, see README_PD.md,
• face detection model, see README_FD.md.
• person-vehicle-bike crossroad detection model, see README_CR.md.

Please find original readme file here.

Models

• Face detection
• Person detection
• Action recognition
• Person-vehicle-bike crossroad detection

Install requirements

1. Install Docker

WARNING Always examine scripts downloaded from the internet before running them locally.

curl -fsSL https://get.docker.com -o get-docker.sh
sudo sh get-docker.sh

2. (optional) Install nvidia-docker plugin

curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | \
  sudo apt-key add -
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | \
  sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt-get update
sudo apt-get install -y nvidia-docker2
sudo pkill -SIGHUP dockerd

3. (optional) Configure proxy settings Create a file /etc/systemd/system/docker.service.d/proxy.conf that adds the proxy environment variables:

[Service]
Environment="HTTP_PROXY=http://proxy.example.com:80/"
Environment="HTTPS_PROXY=https://proxy.example.com:443/"

Flush changes and restart Docker daemon

sudo systemctl daemon-reload
sudo systemctl restart docker

4. Manage Docker as a non-root user

sudo groupadd docker
sudo usermod -aG docker $USER
# Log out and log back in so that your group membership is re-evaluated.

5. (optional) Verify that nvidia-docker is installed correctly

CUDA_VERSION=$(grep -oP '(?<=CUDA Version )(\d+)' /usr/local/cuda/version.txt)
nvidia-docker run --rm nvidia/cuda:${CUDA_VERSION}.0-cudnn7-devel-ubuntu16.04 nvidia-smi

Build instructions

1. Get the code. We will call the directory that you cloned Caffe into $CAFFE_ROOT

git clone https://github.com/opencv/training_toolbox_caffe.git caffe

2. Download openvino package to root directory of the repository

3. Build docker image

./build_docker_image.sh gpu

Run Docker interactive session

NV_GPU=0 nvidia-docker run --rm --name ttcf -it --user=$(id -u):$(id -g) -v <host_path>:<container_path> ttcf:gpu bash

NOTE To run in CPU mode

./build_docker_image.sh cpu
docker run --rm --name ttcf -it --user=$(id -u):$(id -g) -v <host_path>:<container_path> ttcf:cpu bash

And add to all scripts --gpu -1 --image tccf:cpu arguments.

License and Citation

Original Caffe

Caffe is released under the BSD 2-Clause license. The BAIR/BVLC reference models are released for unrestricted use.

Please cite Caffe in your publications if it helps your research:

@article{jia2014caffe,
  Author = {Jia, Yangqing and Shelhamer, Evan and Donahue, Jeff and Karayev, Sergey and Long, Jonathan and Girshick, Ross and Guadarrama, Sergio and Darrell, Trevor},
  Journal = {arXiv preprint arXiv:1408.5093},
  Title = {Caffe: Convolutional Architecture for Fast Feature Embedding},
  Year = {2014}
}

SSD: Single Shot MultiBox Detector

Please cite SSD in your publications if it helps your research:

@inproceedings{liu2016ssd,
  title = {{SSD}: Single Shot MultiBox Detector},
  author = {Liu, Wei and Anguelov, Dragomir and Erhan, Dumitru and Szegedy, Christian and Reed, Scott and Fu, Cheng-Yang and Berg, Alexander C.},
  booktitle = {ECCV},
  year = {2016}
}

AM-Softmax

If you find AM-Softmax useful in your research, please consider to cite:

@article{Wang_2018_amsoftmax,
  title = {Additive Margin Softmax for Face Verification},
  author = {Wang, Feng and Liu, Weiyang and Liu, Haijun and Cheng, Jian},
  journal = {arXiv preprint arXiv:1801.05599},
  year = {2018}
}

WIDERFace dataset

@inproceedings{yang2016wider,
  Author = {Yang, Shuo and Luo, Ping and Loy, Chen Change and Tang, Xiaoou},
  Booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  Title = {WIDER FACE: A Face Detection Benchmark},
  Year = {2016}
}