Skip to content
This repository was archived by the owner on Jan 8, 2026. It is now read-only.

IntelligentSandbox/autovia

Repository files navigation

Autovia

Autovia is a web application for semantic segmentation of road scenes using the CityScapes dataset.

webpage

Installation

Install the Python dependencies:

pip install -r requirements.txt

Install the frontend dependencies:

cd webserver
npm install

Usage

Run both the frontend and backend:

./run.sh

Or run them separately:

# Backend
python backend.py --model_weights_path ./weights/model-bnet-2.pth

# Frontend (in another terminal)
cd webserver
npm start

Training

To train the model yourself, download the CityScapes dataset (requires .edu email). You only need these two files:

  • gtFine_trainvaltest.zip (241MB) - fine annotations for train/val/test sets
  • leftImg8bit_trainvaltest.zip (11GB) - left 8-bit images for train/val/test sets

Extract both to ~/data/cityscapes/ and run:

python train.py

Authors

Ramon Asuncion Batista, Santiago Hernandez, Warren Wang

References

Based on the tutorial code from talhaanwarch/youtube-tutorials.

About

Semantic segmentation autonomous driving web app

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors