Official PyTorch implementation of the IJCB 2023 paper:
"Enhancing 3D-Air Signature by Pen Tip Tail Trajectory Awareness: Dataset and Featuring by Novel Spatio-temporal CNN"
This repository contains benchmark implementations for air signature verification on the T3AAS-v1 dataset.
The T3AAS-v1 (Tip Tail Trajectory Aware Air Signature) dataset represents a significant advancement in 3D air signature verification research.
To obtain access to the T3AAS-v1 dataset, please complete the data access request form:
- Python 3.7+
- CUDA compatible GPU (recommended)
Choose one of the following methods to install dependencies:
conda env create -f environment.yml
conda activate t3aas-envpip install -r requirements.txtAll experiments are managed through the main script:
python main.py [arguments]To view all available arguments and options:
python main.py --helpFor convenience, we provide a sample execution script run.sh.
If you find this implementation useful for your research, please consider citing:
@INPROCEEDINGS{atreya2023enhancing,
author={Atreya, Saurabh and Bora, Maheswar and Mukherjee, Aritra and Das, Abhijit},
booktitle={2023 IEEE International Joint Conference on Biometrics (IJCB)},
title={Enhancing 3D-Air Signature by Pen Tip Tail Trajectory Awareness: Dataset and Featuring by Novel Spatio-temporal CNN},
year={2023},
volume={},
number={},
pages={1-9},
keywords={Three-dimensional displays;Biometrics (access control);Tail;Cameras;Forgery;Trajectory;Convolutional neural networks},
doi={10.1109/IJCB57857.2023.10448666}
}This research is conducted by the Machine Intelligence Group at BITS Pilani, Hyderabad Campus.
This project is licensed under the MIT License - see the LICENSE file for details.