Matches in SemOpenAlex for { <https://semopenalex.org/work/W2548778875> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W2548778875 abstract "Traditional approaches to structured semantic segmentation employ appearance-based classifiers to provide a class-likelihood at each spatial location and then post-process it with Markov Random Fields (MRF) to enforce label smoothness and structure in the output space. The spatial support for such techniques is usually a patch of pixels, which makes the prediction over-smoothed because the borders of objects are not explicitly taken into account. This is further exacerbated by MRF post-processing employing the standard Potts model, which tends to further over-smooth predictions at boundaries. In this paper, we propose a different but related approach: we optimize an energy function finding the optimal combination of small ground truth (GT) tiles from training data over predictions at test time, effectively solving a puzzle. We optimize over a first configuration given by a Convolutional Neural Network (CNN) output." @default.
- W2548778875 created "2016-11-11" @default.
- W2548778875 creator A5005192117 @default.
- W2548778875 creator A5037775056 @default.
- W2548778875 creator A5071561639 @default.
- W2548778875 date "2016-07-01" @default.
- W2548778875 modified "2023-09-28" @default.
- W2548778875 title "Solving structured segmentation of aerial images as puzzles" @default.
- W2548778875 cites W1554544485 @default.
- W2548778875 cites W1909515874 @default.
- W2548778875 cites W1937812750 @default.
- W2548778875 cites W1972562196 @default.
- W2548778875 cites W1974524700 @default.
- W2548778875 cites W1982931780 @default.
- W2548778875 cites W2003339231 @default.
- W2548778875 cites W2102150301 @default.
- W2548778875 cites W2128745377 @default.
- W2548778875 cites W2136064009 @default.
- W2548778875 cites W2141155330 @default.
- W2548778875 cites W2148300333 @default.
- W2548778875 doi "https://doi.org/10.1109/igarss.2016.7729865" @default.
- W2548778875 hasPublicationYear "2016" @default.
- W2548778875 type Work @default.
- W2548778875 sameAs 2548778875 @default.
- W2548778875 citedByCount "0" @default.
- W2548778875 crossrefType "proceedings-article" @default.
- W2548778875 hasAuthorship W2548778875A5005192117 @default.
- W2548778875 hasAuthorship W2548778875A5037775056 @default.
- W2548778875 hasAuthorship W2548778875A5071561639 @default.
- W2548778875 hasConcept C102634674 @default.
- W2548778875 hasConcept C105795698 @default.
- W2548778875 hasConcept C111919701 @default.
- W2548778875 hasConcept C119857082 @default.
- W2548778875 hasConcept C121332964 @default.
- W2548778875 hasConcept C121864883 @default.
- W2548778875 hasConcept C124504099 @default.
- W2548778875 hasConcept C134306372 @default.
- W2548778875 hasConcept C146849305 @default.
- W2548778875 hasConcept C153180895 @default.
- W2548778875 hasConcept C154945302 @default.
- W2548778875 hasConcept C159886148 @default.
- W2548778875 hasConcept C160633673 @default.
- W2548778875 hasConcept C2778045648 @default.
- W2548778875 hasConcept C33923547 @default.
- W2548778875 hasConcept C41008148 @default.
- W2548778875 hasConcept C51329190 @default.
- W2548778875 hasConcept C81363708 @default.
- W2548778875 hasConcept C89600930 @default.
- W2548778875 hasConcept C98045186 @default.
- W2548778875 hasConcept C98763669 @default.
- W2548778875 hasConcept C98925819 @default.
- W2548778875 hasConceptScore W2548778875C102634674 @default.
- W2548778875 hasConceptScore W2548778875C105795698 @default.
- W2548778875 hasConceptScore W2548778875C111919701 @default.
- W2548778875 hasConceptScore W2548778875C119857082 @default.
- W2548778875 hasConceptScore W2548778875C121332964 @default.
- W2548778875 hasConceptScore W2548778875C121864883 @default.
- W2548778875 hasConceptScore W2548778875C124504099 @default.
- W2548778875 hasConceptScore W2548778875C134306372 @default.
- W2548778875 hasConceptScore W2548778875C146849305 @default.
- W2548778875 hasConceptScore W2548778875C153180895 @default.
- W2548778875 hasConceptScore W2548778875C154945302 @default.
- W2548778875 hasConceptScore W2548778875C159886148 @default.
- W2548778875 hasConceptScore W2548778875C160633673 @default.
- W2548778875 hasConceptScore W2548778875C2778045648 @default.
- W2548778875 hasConceptScore W2548778875C33923547 @default.
- W2548778875 hasConceptScore W2548778875C41008148 @default.
- W2548778875 hasConceptScore W2548778875C51329190 @default.
- W2548778875 hasConceptScore W2548778875C81363708 @default.
- W2548778875 hasConceptScore W2548778875C89600930 @default.
- W2548778875 hasConceptScore W2548778875C98045186 @default.
- W2548778875 hasConceptScore W2548778875C98763669 @default.
- W2548778875 hasConceptScore W2548778875C98925819 @default.
- W2548778875 hasLocation W25487788751 @default.
- W2548778875 hasOpenAccess W2548778875 @default.
- W2548778875 hasPrimaryLocation W25487788751 @default.
- W2548778875 hasRelatedWork W2001992768 @default.
- W2548778875 hasRelatedWork W2047139957 @default.
- W2548778875 hasRelatedWork W2083620454 @default.
- W2548778875 hasRelatedWork W2101908237 @default.
- W2548778875 hasRelatedWork W2125517551 @default.
- W2548778875 hasRelatedWork W2128227956 @default.
- W2548778875 hasRelatedWork W2159388312 @default.
- W2548778875 hasRelatedWork W2535890285 @default.
- W2548778875 hasRelatedWork W2601419087 @default.
- W2548778875 hasRelatedWork W2789255933 @default.
- W2548778875 hasRelatedWork W2958925357 @default.
- W2548778875 hasRelatedWork W3015863129 @default.
- W2548778875 hasRelatedWork W3080724905 @default.
- W2548778875 hasRelatedWork W3134627481 @default.
- W2548778875 hasRelatedWork W3211857712 @default.
- W2548778875 hasRelatedWork W78505551 @default.
- W2548778875 hasRelatedWork W8430746 @default.
- W2548778875 hasRelatedWork W2847445408 @default.
- W2548778875 hasRelatedWork W3002526540 @default.
- W2548778875 hasRelatedWork W3002748593 @default.
- W2548778875 isParatext "false" @default.
- W2548778875 isRetracted "false" @default.
- W2548778875 magId "2548778875" @default.
- W2548778875 workType "article" @default.