Search: onr:"swepub:oai:research.chalmers.se:ac6b8a80-1e05-4d17-ad29-5e44c1e2ca9e" > A projected gradien...
Fältnamn | Indikatorer | Metadata |
---|---|---|
000 | 04582naa a2200505 4500 | |
001 | oai:research.chalmers.se:ac6b8a80-1e05-4d17-ad29-5e44c1e2ca9e | |
003 | SwePub | |
008 | 180413s2018 | |||||||||||000 ||eng| | |
009 | oai:lup.lub.lu.se:bf660148-877e-4d0e-9fa5-0bcbdd9b2e3d | |
024 | 7 | a https://doi.org/10.1007/978-3-319-78199-0_372 DOI |
024 | 7 | a https://research.chalmers.se/publication/5018762 URI |
024 | 7 | a https://lup.lub.lu.se/record/bf660148-877e-4d0e-9fa5-0bcbdd9b2e3d2 URI |
040 | a (SwePub)cthd (SwePub)lu | |
041 | a engb eng | |
042 | 9 SwePub | |
072 | 7 | a kon2 swepub-publicationtype |
072 | 7 | a ref2 swepub-contenttype |
100 | 1 | a Larsson, Måns,d 1989u Chalmers University of Technology4 aut0 (Swepub:cth)lmans |
245 | 1 0 | a A projected gradient descent method for crf inference allowing end-to-end training of arbitrary pairwise potentials |
264 | c 2018-03-22 | |
264 | 1 | a Cham :b Springer International Publishing,c 2018 |
520 | a Are we using the right potential functions in the Conditional Random Field models that are popular in the Vision community? Semantic segmentation and other pixel-level labelling tasks have made significant progress recently due to the deep learning paradigm. However, most state-of-the-art structured prediction methods also include a random field model with a hand-crafted Gaussian potential to model spatial priors, label consistencies and feature-based image conditioning. In this paper, we challenge this view by developing a new inference and learning framework which can learn pairwise CRF potentials restricted only by their dependence on the image pixel values and the size of the support. Both standard spatial and high-dimensional bilateral kernels are considered. Our framework is based on the observation that CRF inference can be achieved via projected gradient descent and consequently, can easily be integrated in deep neural networks to allow for end-to-end training. It is empirically demonstrated that such learned potentials can improve segmentation accuracy and that certain label class interactions are indeed better modelled by a non-Gaussian potential. In addition, we compare our inference method to the commonly used mean-field algorithm. Our framework is evaluated on several public benchmarks for semantic segmentation with improved performance compared to previous state-of-the-art CNN+CRF models. | |
650 | 7 | a NATURVETENSKAPx Data- och informationsvetenskapx Annan data- och informationsvetenskap0 (SwePub)102992 hsv//swe |
650 | 7 | a NATURAL SCIENCESx Computer and Information Sciencesx Other Computer and Information Science0 (SwePub)102992 hsv//eng |
650 | 7 | a NATURVETENSKAPx Data- och informationsvetenskapx Bioinformatik0 (SwePub)102032 hsv//swe |
650 | 7 | a NATURAL SCIENCESx Computer and Information Sciencesx Bioinformatics0 (SwePub)102032 hsv//eng |
650 | 7 | a NATURVETENSKAPx Data- och informationsvetenskapx Datorseende och robotik0 (SwePub)102072 hsv//swe |
650 | 7 | a NATURAL SCIENCESx Computer and Information Sciencesx Computer Vision and Robotics0 (SwePub)102072 hsv//eng |
653 | a Convolutional neural networks | |
653 | a Segmentation | |
653 | a Conditional random fields | |
653 | a Conditional random fields | |
653 | a Convolutional neural networks | |
653 | a Segmentation | |
700 | 1 | a Arnab, Anuragu University Of Oxford,University of Oxford4 aut |
700 | 1 | a Kahl, Fredrik,d 1972u Chalmers University of Technology,Lund University,Lunds universitet,Matematik LTH,Matematikcentrum,Institutioner vid LTH,Lunds Tekniska Högskola,Mathematics (Faculty of Engineering),Centre for Mathematical Sciences,Departments at LTH,Faculty of Engineering, LTH4 aut0 (Swepub:lu)math-fka |
700 | 1 | a Zheng, Shuaiu University Of Oxford,University of Oxford4 aut |
700 | 1 | a Torr, Philip H.S.u University Of Oxford,University of Oxford4 aut |
710 | 2 | a Chalmers University of Technologyb University Of Oxford4 org |
773 | 0 | t Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)d Cham : Springer International Publishingg 10746 LNCS, s. 564-579q 10746 LNCS<564-579x 1611-3349x 0302-9743z 9783319781983 |
856 | 4 | u http://arxiv.org/pdf/1701.06805 |
856 | 4 | u http://dx.doi.org/10.1007/978-3-319-78199-0_37y FULLTEXT |
856 | 4 8 | u https://doi.org/10.1007/978-3-319-78199-0_37 |
856 | 4 8 | u https://research.chalmers.se/publication/501876 |
856 | 4 8 | u https://lup.lub.lu.se/record/bf660148-877e-4d0e-9fa5-0bcbdd9b2e3d |
Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.