A semantic Segmentation model used to identify road surfaces for self-driving car applications. Implemented in Tensorflow and trained on the Kitti Road Dataset.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.gitignore
README.md
architectures.py
base.py
data_processing.py
image_processing.py
shadow_pattern.jpg
train.py
video_feed.py
viz.py

README.md

Semantic Segmentation on the Kitti dataset

A deep neural network is trained for the task of semantic segmentation of images.

...

Data

The data used is the The KITTI road data.

The specific dataset file used can be downloaded with this link

Preparing the data

Notes coming soon.

Model architecture

...

Results

...