Object Extraction Dataset

dataset overview

Introduction

This Object Extraction newly collected by us contains 10183 images with groundtruth segmentation masks. We selected the images from the PASCAL[1], iCoseg[2], Internet [3] dataset as well as other data (most of them are about people and clothes) from the web. We randomly split the dataset with 8230 images for training and 1953 images for testing.

Download The DataSet

  • image.zip: It contains the original images.
  • groundtruth.zip: It contains the groundtruth segmentation masks.
  • train.list: It contains the file names for the train set
  • test.list: It contains the file names for the test set
  • References

    [1] M. Everingham, L. Van Gool, C. Williams, J. Winn, and A. Zisserman, The Pascal Visual Object Classes (VOC) Challenge, In Intl J. of Computer Vision, 88:303-338,2010.

    [2] D. Batra, A. Kowdle, D. Parikh, J. Luo, and T. Chen, iCoseg: Interactive Co-segmentation with Intelligent Scribble Guidance, In CVPR, 2010.

    [3] M. Rubinstein, A. Joulin, J. Kopf, and C. Liu, Unsupervised Joint Object Discovery and Segmentation in Internet Images, In CVPR, 2013.