%0 Journal Article %1 zheng2015conditional %A Zheng, Shuai %A Jayasumana, Sadeep %A Romera-Paredes, Bernardino %A Vineet, Vibhav %A Su, Zhizhong %A Du, Dalong %A Huang, Chang %A Torr, Philip H. S. %D 2015 %J CoRR %K %T Conditional Random Fields as Recurrent Neural Networks %U https://arxiv.org/pdf/1502.03240.pdf %V abs/1502.03240 %X Pixel-level labelling tasks, such as semantic segmentation, play a central role in image understanding. Recent approaches have attempted to harness the capabilities of deep learning techniques for image recognition to tackle pixel-level labelling tasks. One central issue in this methodology is the limited capacity of deep learning techniques to delineate visual objects. To solve this problem, we introduce a new form of convolutional neural network that combines the strengths of Convolutional Neural Networks (CNNs) and Conditional Random Fields (CRFs)-based probabilistic graphical modelling. To this end, we formulate mean-field approximate inference for the Conditional Random Fields with Gaussian pairwise potentials as Recurrent Neural Networks. This network, called CRF-RNN, is then plugged in as a part of a CNN to obtain a deep network that has desirable properties of both CNNs and CRFs. Importantly, our system fully integrates CRF modelling with CNNs, making it possible to train the whole deep network end-to-end with the usual back-propagation algorithm, avoiding offline post-processing methods for object delineation. We apply the proposed method to the problem of semantic image segmentation, obtaining top results on the challenging Pascal VOC 2012 segmentation benchmark.

@article{zheng2015conditional, abstract = {Pixel-level labelling tasks, such as semantic segmentation, play a central role in image understanding. Recent approaches have attempted to harness the capabilities of deep learning techniques for image recognition to tackle pixel-level labelling tasks. One central issue in this methodology is the limited capacity of deep learning techniques to delineate visual objects. To solve this problem, we introduce a new form of convolutional neural network that combines the strengths of Convolutional Neural Networks (CNNs) and Conditional Random Fields (CRFs)-based probabilistic graphical modelling. To this end, we formulate mean-field approximate inference for the Conditional Random Fields with Gaussian pairwise potentials as Recurrent Neural Networks. This network, called CRF-RNN, is then plugged in as a part of a CNN to obtain a deep network that has desirable properties of both CNNs and CRFs. Importantly, our system fully integrates CRF modelling with CNNs, making it possible to train the whole deep network end-to-end with the usual back-propagation algorithm, avoiding offline post-processing methods for object delineation. We apply the proposed method to the problem of semantic image segmentation, obtaining top results on the challenging Pascal VOC 2012 segmentation benchmark.}, added-at = {2017-02-26T17:58:42.000+0100}, author = {Zheng, Shuai and Jayasumana, Sadeep and Romera-Paredes, Bernardino and Vineet, Vibhav and Su, Zhizhong and Du, Dalong and Huang, Chang and Torr, Philip H. S.}, biburl = {https://www.bibsonomy.org/bibtex/27b1185aa9eda013f8c8fe351fd75fd17/nosebrain}, interhash = {0005887723a4b8a07c67a94ea7a56104}, intrahash = {7b1185aa9eda013f8c8fe351fd75fd17}, journal = {CoRR}, keywords = {}, timestamp = {2017-02-26T17:58:42.000+0100}, title = {Conditional Random Fields as Recurrent Neural Networks}, url = {https://arxiv.org/pdf/1502.03240.pdf}, volume = {abs/1502.03240}, year = 2015 }