Dim P. Papadopoulos, Jasper R. R. Uijlings, Frank Keller, Vittorio Ferrari

Overview

Welcome to this release of the center-click annotations on PASCAL VOC 2007 trainval images. We collected center-click annotations on Amazon Mechanical Turk (AMT) for all 20 classes of the whole trainval set of PASCAL VOC 2007 [2]. Each image was annotated with a click by two different annotators for each class present in the image. This results in 14,612 clicks in total for the 5,011 trainval images.
(see [1] for more details about the crowd-sourcing annotation protocol).

We also provide the object localizations produced by our proposed center-click supervision approach described in our CVPR 2017 paper [1].

Downloads

The PASCAL VOC 2007 trainval images can be found here.

Explore

Below you can see a sample of the center-click annotations and our output object localizations on PASCAL VOC 2007 trainval images.

References

1. Training object class detectors with click supervision
   D. P. Papadopoulos, J. R. R. Uijlings, F. Keller and V. Ferrari,
   In Computer Vision and Pattern Recognition (CVPR), 2017.
   
2. The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results
   M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn and A. Zisserman