SkipNet-FloorPlanGen

This the official repo of report for ICCV 2023 1st Computer Vision Aided Architectural Design(CVAAD) Workshop

Report Link

Table of Contents

1. Environment Setup
2. Dataset Preprocessing
3. Training
4. Evaluation and Submission
5. Inference Visualization

Environment Setup

Create a Conda environment:

conda create -n plangen python==3.9.0
conda activate plangen
conda install pytorch==2.0.0 torchvision==0.15.0 torchaudio==2.0.0 pytorch-cuda=11.8 -c pytorch -c nvidia
pip install -r requirements.txt

Dataset Preprocessing

We utilize mobileSAM to preprocess the boundary images and organize the dataset into train/val/test sets for training.

Below is the link to the actual preprocessed dataset used for training:

Dataset Link

Training

Run the following command for training:

python train.py

We provide our training model's weight : model_weight

Evaluation and Submission

Run the following command for evaluation and submission:

python submission.py

Inference Visualization

Run the inference_visualization.ipynb Jupyter Notebook for inference visualization.