Repository files navigation

This project utilizes a modified MobileNet in company with theSSD framework to achieve a robust and fast detection of hand location and orientation.Our implementation is based onthe PyTorch version of SSD andMobileNet.

1. Preparation
2. Training
3. Evaluation
4. Citation

1. Due to some compatibility issues, we recommend to install PyTorch 0.3.0 and Python 3.6.8, which our project currently supports.

2. Get the code.

   git clone https://github.com/yangli18/hand_detection.git

3. Downloadthe Oxford hand dataset and create the LMDB file for the training data.

   sh data/scripts/Oxford_hand_dataset.sh

4. Compile the NMS code (fromruotianluo/pytorch-faster-rcnn).

   sh layers/src/make.sh

Train the detection model on the Oxford hand dataset.

python train.py2>&1| tee log/train.log

• Trained with data augmentation v2, our detection model reaches over86% average precision (AP) on the Oxford hand dataset. We provide the trained model in theweights dir.
• For the convenience of training, a pre-trained MobileNet is put in theweights dir.You can also download it fromhere.

1. Evaluate the trained detection model.

   python eval.py --trained_model weights/ssd_new_mobilenet_FFA.pth --version ssd_new_mobilenet_FFA

   • Note: For a fair comparison, the evaluation code of the Oxford hand dataset should be used to get the exact mAP (mean Average Precision) of hand detection.The above command should return a similar result, but not exactly the same.

2. Evaluate the average detection time.

   python eval_speed.py --trained_model weights/ssd_new_mobilenet_FFA.pth --version ssd_new_mobilenet_FFA

If you find our code useful, please cite our paper.

@inproceedings{yang2018light,  title={A Light CNN based Method for Hand Detection and Orientation Estimation},  author={Yang, Li and Qi, Zhi and Liu, Zeheng and Zhou, Shanshan and Zhang, Yang and Liu, Hao and Wu, Jianhui and Shi, Longxing},  booktitle={2018 24th International Conference on Pattern Recognition (ICPR)},  pages={2050--2055},  year={2018},  organization={IEEE}}@article{yang2019embedded,  title={An embedded implementation of CNN-based hand detection and orientation estimation algorithm},  author={Yang, Li and Qi, Zhi and Liu, Zeheng and Liu, Hao and Ling, Ming and Shi, Longxing and Liu, Xinning},  journal={Machine Vision and Applications},  volume={30},  number={6},  pages={1071--1082},  year={2019},  publisher={Springer}}