Skip to content

VisionLearningGroup/SANE

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This repository implements:

Bryan A. Plummer*, Mariya I. Vasileva*, Vitali Petsiuk, Kate Saenko, David Forsyth.

Why do These Match? Explaining the Behavior of Image Similarity Models. ECCV, 2020.

Environment

This code was tested with Python 3.6 and Pytorch 1.4.

Preparation

Download data and unzip it in ./data

Saliency Map Test

To reproduce results from Table 1 in our paper you can use the saliency.py script in tools, e.g.,

python ./tools/saliency.py --fixed_ref --dataset polyvore_outfits --method rise

Training Attribute Classifier

  1. Cache saliency maps (used in Eq.2 of our paper)
python ./tools/saliency.py --fixed_ref --dataset polyvore_oufits --method rise --split train
  1. Train the model
python ./tools/train_attribute_classifier.py --fixed_ref --dataset polyvore_outfits --method rise

Citation

If you find our code useful please consider citing:

@InProceedings{plummerSimilarityExplanations2020,
     author={Bryan A. Plummer and Mariya I. Vasileva and Vitali Petsiuk and Kate Saenko and David Forsyth},
     title={Why do These Match? Explaining the Behavior of Image Similarity Models},
     booktitle = {The European Conference on Computer Vision (ECCV)},
     year = {2020}
}

About

Pytorch implementation of our ECCV paper

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages