Python scripts for performing stereo depth estimation using the MobileStereoNet model in Tensorflow Lite.

Stereo depth estimation on the cones images from the Middlebury dataset (https://vision.middlebury.edu/stereo/data/scenes2003/)

• Check the requirements.txt file. Additionally, pafy and youtube-dl are required for youtube video inference.
• DrivingStereo dataset, ONLY for the driving_sereo_test.pyscript. Link: https://drivingstereo-dataset.github.io/

pip install -r requirements.txt
pip install pafy youtube-dl

The original models was converted to different formats (including .tflite) by PINTO0309, the models can be found in his repository.

The Pytorch pretrained model was taken from the original repository.

• Image inference:

python image_depth_estimation.py 

• Video inference:

python video_depth_estimation.py

• DrivingStereo dataset inference:

python driving_sereo_test.py

• MobileStereoNet model: https://github.com/cogsys-tuebingen/mobilestereonet
• PINTO0309's model zoo: https://github.com/PINTO0309/PINTO_model_zoo
• PINTO0309's model conversion tool: https://github.com/PINTO0309/openvino2tensorflow
• DrivingStereo dataset: https://drivingstereo-dataset.github.io/
• Original paper: https://arxiv.org/pdf/2108.09770.pdf