Convolutional CRFs for semantic segmentation
View / Open Files
Publication Date
2020-01-01Journal Title
30th British Machine Vision Conference 2019, BMVC 2019
Conference Name
BMVC
Type
Conference Object
This Version
AM
Metadata
Show full item recordCitation
Teichmann, M., & Cipolla, R. (2020). Convolutional CRFs for semantic segmentation. 30th British Machine Vision Conference 2019, BMVC 2019 https://doi.org/10.17863/CAM.42064
Abstract
For the challenging semantic image segmentation task the best performing models
have traditionally combined the structured modelling capabilities of Conditional Random
Fields (CRFs) with the feature extraction power of CNNs. In more recent works however,
CRF post-processing has fallen out of favour. We argue that this is mainly due to the slow
training and inference speeds of CRFs, as well as the difficulty of learning the internal
CRF parameters. To overcome both issues we propose to add the assumption of conditional
independence to the framework of fully-connected CRFs. This allows us to reformulate the
inference in terms of convolutions, which can be implemented highly efficiently on GPUs.
Doing so speeds up inference and training by two orders of magnitude. All parameters of
the convolutional CRFs can easily be optimized using backpropagation. Towards the goal
of facilitating further CRF research we have made our implementations publicly available.
Identifiers
External DOI: https://doi.org/10.17863/CAM.42064
This record's URL: https://www.repository.cam.ac.uk/handle/1810/294983
Rights
All rights reserved
Licence:
http://www.rioxx.net/licenses/all-rights-reserved
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.