README.md

KITTI DeepLab

Inference script and frozen inference graph with fine tuned weights for semantic segmentation on images from the KITTI dataset with TensorFlow. While the weights provided by the DeepLab authors reach an mIoU of 44% on the KITTI validation set, the fine-tuned weights reach an mIoU of 72.73 %.

Example output

Original image

Segmentation Result with original graph

Segmentation Result with fine-tuned graph

Getting Started

Clone the repository: git clone https://github.com/hiwad-aziz/kitti_deeplab.git

Download the frozen inference graph: wget --no-check-certificate -r 'https://docs.google.com/uc?export=download&id=1ynVTP5CjMui2J3j2JLJDii8oC0ef_WpL' -O frozen_inference_graph.zip

Run the script: python inference.py path_to_frozen_inference_graph.pb path_to_image_folder/

References

DeepLab

https://github.com/tensorflow/models/tree/master/research/deeplab

KITTI dataset

http://www.cvlibs.net/datasets/kitti/