Skip to content
Code for AutoDispNet (ICCV 2019)
Python Shell
Branch: master
Clone or download
Latest commit 0c08842 Dec 20, 2019
Type Name Latest commit message Commit time
Failed to load latest commit information.
controller Remove ununsed module Dec 20, 2019
nets Fix import for named_schedule Dec 20, 2019
operations Remove ununsed module Dec 20, 2019
.gitignore Remove ununsed module Dec 20, 2019
LICENSE Add license Dec 17, 2019
NOTICE Add notice Dec 20, 2019 Update Dec 20, 2019 Fixed dependecies Dec 20, 2019
teaser-1.png Added teaser Dec 20, 2019


Code accompanying the paper: AutoDispNet: Improving Disparity Estimation with AutoML (ICCV 2019). Parts of this codebase is inspired from DARTS.

Note: We provide deployment code only.


Running networks

  • Change your directory to the network directory (autodispnet/nets).

  • Download pre-trained weights with Pre-trained weights are provided for networks trained on FlyingThings (CSS, css) and fine-tuned on KITTI (CSS-KITTI). css is a network with smaller memory footprint (see paper for details).

  • Go to a network directory (Eg: autodispnet/nets/CSS) and use the following command to test the network on an image pair:

    python3 eval image0_path image1_path out_dir

  • The output is stored in a binary format with .float3 extension (Information on reading the output is here).


If you use the code or parts of it in your research, you should cite the aforementioned paper:

  author       = "T. Saikia and Y. Marrakchi and A. Zela and F. Hutter and T. Brox",
  title        = "AutoDispNet: Improving Disparity Estimation With AutoML",
  booktitle    = "IEEE International Conference on Computer Vision (ICCV)",
  month        = "October",
  year         = "2019",
  url          = ""


Tonmoy Saikia (

You can’t perform that action at this time.