Awesome

Introduction

This is the pytorch implementation of the paper "Towards Context-Aware Interaction Recognition for Visual Relationship Detection"

If you use this code in your research, please cite our paper:

@InProceedings{Zhuang_2017_ICCV,
author = {Zhuang, Bohan and Liu, Lingqiao and Shen, Chunhua and Reid, Ian},
title = {Towards Context-Aware Interaction Recognition for Visual Relationship Detection},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {Oct},
year = {2017}
}

Dataset and Evaluation Metrics

Please download the VRD dataset and the corresponding evaluation metrics from https://github.com/Prof-Lu-Cewu/Visual-Relationship-Detection

Note that we first extract the data into individual files for training/testing.

Code

The code are written using Pytorch.

utils.py: provide necessary functions
new_layers.py: provide self-defined layers
train.py: main file, implementing training and testing
config.yaml: define the necessary hyperparameters (e.g., data directory, GPU), please modify this file
model.py: define network structures

If you want to evaluate the context-aware model independently without attention, find the code provided in ./no_attention subfolder.

Training

python train.py

Testing

Follow the evaluation instructions in "https://github.com/Prof-Lu-Cewu/Visual-Relationship-Detection". Please extract the features by yourself after training and evaluate using relationship_phrase_detection.m and predicate_detection.m, respectively.

Copyright

** This code is for non-commercial purposes only. For commerical purposes, please contact Chunhua Shen chhshen@gmail.com **

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.