Skip to content

Code that implements paper "End-to-End Instance Segmentation with Recurrent Attention"

License

Notifications You must be signed in to change notification settings

renmengye/rec-attend-public

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

rec-attend-public

Code that implements paper "End-to-End Instance Segmentation with Recurrent Attention".

Dependencies

  • Python 2.7
  • TensorFlow 0.12 (not compatible with TensorFlow 1.0)
  • OpenCV
  • NumPy
  • SciPy
  • PyYaml
  • hdf5 and H5Py
  • tqdm
  • Pillow (required by cityscapes evaluation)

Installation

Compile Hungarian matching module

./hungarian_build.sh

CVPPP Experiments

First modify setup_cvppp.sh with your dataset folder paths.

./setup_cvppp.sh

Run experiments:

./run_cvppp.sh

KITTI Experiments

First modify setup_kitti.sh with your dataset folder paths.

./setup_kitti.sh

Run experiments:

./run_cvppp.sh

Cityscapes Experiments

First modify setup_cityscapes.sh with your dataset folder paths.

./setup_cityscapes.sh

Run experiments:

./run_cityscapes.sh

Citation

If you use our code, please consider cite the following: End-to-End Instance Segmentation with Recurrent Attention. Mengye Ren, Richard S. Zemel. CVPR 2017.

@inproceedings{ren17recattend,
  author    = {Mengye Ren and Richard S. Zemel},
  title     = {End-to-End Instance Segmentation with Recurrent Attention},
  booktitle = {CVPR},
  year      = {2017}
}

About

Code that implements paper "End-to-End Instance Segmentation with Recurrent Attention"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published