https://github.com/Yevkuzn/semodepth

Semi-supervised monocular depth map prediction
https://github.com/Yevkuzn/semodepth

deep-learning depth-estimation monocular-depth

Last synced: 6 days ago
JSON representation

Semi-supervised monocular depth map prediction

• Host: GitHub
• URL: https://github.com/Yevkuzn/semodepth
• Owner: Yevkuzn
• License: gpl-3.0
• Created: 2018-04-19T09:34:30.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-08-05T23:58:26.000Z (over 6 years ago)
• Last Synced: 2024-08-02T08:10:08.710Z (3 months ago)
• Topics: deep-learning, depth-estimation, monocular-depth
• Language: Python
• Size: 30.3 KB
• Stars: 72
• Watchers: 4
• Forks: 15
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-depth-estimation - Code
• awesome-depth - 3

README

        
# Semi-Supervised Deep Learning for Monocular Depth Map Prediction

This repository contains code for the depth estimation system as described in

**[Semi-Supervised Deep Learning for Monocular Depth Map Prediction, CVPR 2017](http://openaccess.thecvf.com/content_cvpr_2017/papers/Kuznietsov_Semi-Supervised_Deep_Learning_CVPR_2017_paper.pdf)**

By Yevhen Kuznietsov, Jörg Stückler, Bastian Leibe at Computer Vision Group, RWTH Aachen University

## Presentation video



## Prerequisite

In order to run the code, your setup has to meet the following requirements (tested versions in parentheses. Other versions might work as well):

* Python (2.7.14)

  * SciPy (1.0.0)

* TensorFlow (1.4.0)

  * CUDA (9.1.x)

  * cuDNN (7.0.x)

* GPU compatible with CUDAv9.1

### Running the system

1.  Download network model [here](https://www.vision.rwth-aachen.de/media/papers/best_model.tgz)

2.  Create a file, containing input-output file paths in each line. The format to be used for each line is `input_path/input.png,output_path/output.mat`. Example can be found at [filenames.txt](filenames.txt)

3.  Edit the [config file](/inference/config.py), set all the paths.

4.  Run the system: `python %PROJ_DIR%/inference/produce_predictions.py` 

## Remarks

* While the model may work with other datasets, this code is only supposed to be run with input resolution of [KITTI](http://www.cvlibs.net/datasets/kitti/) dataset

* The code was tested with 6GB NVIDIA GeForce GTX 980 Ti. It should also be possible to run it with less GPU RAM. Running on CPU may require code modifications.

* The metric depth map predictions, as well as the inputs and the generated depth ground truth for 'Eigen' test set are available [here](https://www.vision.rwth-aachen.de/publication/00150/)

If you have any issues or questions about the code, you can contact [me](mailto:[email protected]) or [my alter ego](mailto:[email protected])

## Citing

If you find the depth estimation model useful in your research, please consider citing:

    @InProceedings{Kuznietsov_2017_CVPR,

        author = {Kuznietsov, Yevhen and Stuckler, Jorg and Leibe, Bastian},

        title = {Semi-Supervised Deep Learning for Monocular Depth Map Prediction},

        booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},

        month = {July},

        year = {2017}

    }