Home
Uni-Logo
 

Object segmentation by alignment of poselet activations to image contours

Thomas Brox, L. Bourdev, S. Maji, J. Malik
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2011
Abstract: In this paper, we propose techniques to make use of two complementary bottom-up features, image edges and texture patches, to guide top-down object segmentation towards higher precision. We build upon the part-based poselet detector, which can predict masks for numerous parts of an object. For this purpose we extend poselets to 19 other categories apart from person. We non-rigidly align these part detections to potential object contours in the image, both to increase the precision of the predicted object mask and to sort out false positives. We spatially aggregate object information via a variational smoothing technique while ensuring that object regions do not overlap. Finally, we propose to refine the segmentation based on self-similarity defined on small image patches. We obtain competitive results on the challenging Pascal VOC benchmark. On four classes we achieve the best numbers to-date.


Other associated files : brox_cvpr11.pdf [843KB]  

Images and movies

 

BibTex reference

@InProceedings{Bro11b,
  author       = "T. Brox and L. Bourdev and S. Maji and J. Malik",
  title        = "Object segmentation by alignment of poselet activations to image contours",
  booktitle    = "IEEE International Conference on Computer Vision and Pattern Recognition (CVPR)",
  month        = "Jun.",
  year         = "2011",
  url          = "http://lmb.informatik.uni-freiburg.de/Publications/2011/Bro11b"
}

Other publications in the database