Skip to content

asanakoy/deep_unsupervised_posets

master
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 
 
 

Deep Unsupervised Similarity Learning Using Partially Ordered Sets (CVPR 2017)

Accepted at CVPR 2017
"Deep Unsupervised Similarity Learning Using Partially Ordered Sets" Miguel A. Bautista* , Artsiom Sanakoyeu* , Björn Ommer.


Tensorflow models for Olympic Sports dataset trained with our approach

All models were trained from scratch without Imagenet pretraining and without any supervision.

  1. The model trained on all frames from Olympic sports dataset: olympic_sports_all_cat_convnet_scratch_strip.ckpt
  2. Using the same method we finetuned the previous model for each sport independently w/o any supervision (we again used only grouping and posets that we build without GT information).
    Single models for each sport: olympic_sports_models_from_scratch

Requirements

  • Python 2.7
  • Tensorflow r1.*

Example

Example how to load models: example_load_networks.ipynb.


If you find this code or data useful for your research, please cite

@inproceedings{UnsupSimPosets2017,
  title={Deep Unsupervised Similarity Learning using Partially Ordered Sets}
  author={Bautista, Miguel A and Sanakoyeu, Artsiom and Ommer, Bj{\"o}rn},
  booktitle={The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2017}
}

About

Deep Unsupervised Similarity Learning using Partially Ordered Sets (CVPR17)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published