Timezone: »
Most state-of-the-art techniques for multi-class image segmentation and labeling use conditional random fields defined over pixels or image regions. While region-level models often feature dense pairwise connectivity, pixel-level models are considerably larger and have only permitted sparse graph structures. In this paper, we consider fully connected CRF models defined on the complete set of pixels in an image. The resulting graphs have billions of edges, making traditional inference algorithms impractical. Our main contribution is a highly efficient approximate inference algorithm for fully connected CRF models in which the pairwise edge potentials are defined by linear combinations of Gaussian kernels. Our algorithm can approximately minimize fully connected models on tens of thousands of variables in a fraction of a second. Quantitative and qualitative results on the MSRC-21 and PASCAL VOC 2010 datasets demonstrate that full pairwise connectivity at the pixel level produces significantly more accurate segmentations and pixel-level label assignments.
Author Information
Philipp Krähenbühl (UT Austin)
Vladlen Koltun (Adobe Research)
Related Events (a corresponding poster, oral, or spotlight)
-
2011 Oral: Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials »
Wed. Dec 14th 10:10 -- 10:30 AM Room
More from the Same Authors
-
2021 Poster: Multimodal Virtual Point 3D Detection »
Tianwei Yin · Xingyi Zhou · Philipp Krähenbühl -
2011 Poster: Nonlinear Inverse Reinforcement Learning with Gaussian Processes »
Sergey Levine · Zoran Popovic · Vladlen Koltun -
2010 Poster: Feature Construction for Inverse Reinforcement Learning »
Sergey Levine · Zoran Popovic · Vladlen Koltun