PA
Pamasi/sgg_av
Scene Graph Generation in Autonomous Driving: a neuro-symbolic approach
Scene Graph Generation in Autonomous Driving: a neuro-symbolic approach
Table of Contents
About The Project
The master thesis explores the usage of a Neuro-Symbolic Relational TRansformer (Nesy-RelTR) in the context of Visual Relationship Detection for Autonomous Driving scenarios.
Distributed under the Creative Commons Attribution Non-commercial No Derivatives.
If you find this project useful in your research, please consider cite:
@mastersthesis{sgg_av,
author = {Paolo E.I. Dimasi},
title = {Scene Graph Generation in Autonomous Driving: a Neuro-symbolic Approach},
school = {Politecnico di Torino},
year = {2023},
address = {Turin, Italy},
url = {http://webthesis.biblio.polito.it/id/eprint/29354}
}Getting Started
Prerequisites
Python 3.10+
Installation
- Clone the repo
git clone https://github.com/Pamasi/sgg_av.git
- Create an Anaconda environment
conda create -n sgg_av_env python=3.10.12 conda activate sgg_av_env python -m pip install -r requirements.txt
- Generate Traffic Genome dataset (folder coco_traffic_genome_v2) and its extension with Visual Genome (folder coco_mix_dataset_v2)
. generate_tg.sh
WARNING
Torchviz is not compatible with RelTR do not install it.
Create RDF Knowledge graph from pandaset
python generate_kg/pandaset2RDF.py -f <pandaset-metadata-folder> -d <dataset-name> Usage
Knowledge Graph Embedding Generation
python train_kge.pyRelTR: KGE
python train_reltr.py --enable_kge --kge_path <kge_ckpt_path>RelTR: LTN
python train_reltr.py --enable_ltnInference
python inference.py --in_xywh --resume $model_path --img_path <image_path> --conf <confidence_score> --inf_dir <output-folder> Contact
Email: paolo.dimasi@outlook.com