Home

Awesome

Pixel Objectness with Bilinear Filtering

This code is mostly adapted from Pixel Objectness and Fast Bilateral Solver. This uses Bilateral filtering as quick post processing step to clean the object masks produced by the Pixel Objectness Code. For example,

example

Pixel Objectness

The following repository contains pretrained model for pixel objectness.

Please visit our project page for the paper and visual results.

If you use this in your research, please cite the following paper:

@article{pixelobjectness,
  Author = {Jain, Suyog and Xiong, Bo and Grauman, Kristen},
  Journal = {arXiv preprint arXiv:1701.05349},
  Title = {Pixel Objectness},
  Year = {2017}
}

These models are freely available for research and academic purposes. However it's patent pending, so please contact us for any commercial use.

Using the pretrained models:

Download pretrained model from here

This model is trained using Deeplab-v1 caffe library. Please cite [1] and [2] if you use the code.

Visualizing the results:

After execution demo.py will store pixel objectness results as matlab files.

Please refer to show_results.m to see how to visualize and extract foreground masks.

Please cite these too if you use the code:

[1] Caffe:

@article{jia2014caffe,
Author = {Jia, Yangqing and Shelhamer, Evan and Donahue, Jeff and Karayev, Sergey and Long, Jonathan and Girshick, Ross and Guadarrama, Sergio and Darrell, Trevor},
Journal = {arXiv preprint arXiv:1408.5093},
Title = {Caffe: Convolutional Architecture for Fast Feature Embedding},
Year = {2014}
}

[2] Deeplab-v1:

@inproceedings{chen14semantic,
title={Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs},
author={Liang-Chieh Chen and George Papandreou and Iasonas Kokkinos and Kevin Murphy and Alan L Yuille},
booktitle={ICLR},
url={http://arxiv.org/abs/1412.7062},
year={2015}
}