A Quality of Life Perspective for Bangkok City
MIT LicenseUrban development hinges on improving the Quality of Life (QOL) for city inhabitants. Traditionally, QOL assessments rely heavily on questionnaire surveys, which, while informative, can be costly and time-consuming. Leveraging the rapid advancements in Artificial Intelligence, this work introduces an innovative approach to automatically extract mobility indicatorsβkey components of QOL evaluationsβusing Semantic Segmentation and Object Recognition techniques. Our method not only enhances the accuracy of transportation mobility assessments but also significantly reduces the data collection costs associated with QOL evaluations.
Source: Bangkok Garden
git clone https://github.com/kaopanboonyuen/QOL-TransportAI.git
cd QOL-TransportAI
pip install -r requirements.txt
Edit config.yaml
to suit your dataset.
python preprocess.py --data_path /path/to/data --output_path /path/to/output
python train.py --config config.yaml
python evaluate.py --model_path /path/to/model --test_data /path/to/test_data
python inference.py --image_path /path/to/image.png --output_path /path/to/output.png
TransportationMobilityFactorExtraction/
βββ data/ # Datasets and preprocessing scripts
βββ models/ # Model architectures and training scripts
βββ config.yaml # Configuration file
βββ train.py # Training script
βββ evaluate.py # Evaluation script
βββ inference.py # Inference script
βββ README.md # Project documentation
@inproceedings{kijsirikul2019transportation,
title={Transportation mobility factor extraction using image recognition techniques},
author={Kijsirikul, Boonserm and Panboonyuen, Teerapong and Iwahori, Yuji and Hayashi, Yoshitsugu and Vateekul, Peerapon and Achariyaviriya, Witsarut},
booktitle={2019 First International Conference on Smart Technology \& Urban Development (STUD)},
pages={1--7},
year={2019},
organization={IEEE}
}
This project is licensed under the MIT License. See the LICENSE file for details.
This project was made possible by the contributions of our dedicated team and the support of the research community. Special thanks to the STUD 2019 reviewers for their feedback.