Matches in SemOpenAlex for { <https://semopenalex.org/work/W2892635580> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2892635580 abstract "In this paper, we present confidence inference approachin an unsupervised way in stereo matching. Deep Neu-ral Networks (DNNs) have recently been achieving state-of-the-art performance. However, it is often hard to tellwhether the trained model was making sensible predictionsor just guessing at random. To address this problem, westart from a probabilistic interpretation of theL1loss usedin stereo matching, which inherently assumes an indepen-dent and identical (aka i.i.d.) Laplacian distribution. Weshow that with the newly introduced dense confidence map,the identical assumption is relaxed. Intuitively, the vari-ance in the Laplacian distribution is large for low confidentpixels while small for high-confidence pixels. In practice,the network learns toattenuatelow-confidence pixels (e.g.,noisy input, occlusions, featureless regions) andfocusonhigh-confidence pixels. Moreover, it can be observed fromexperiments that the focused learning is very helpful in find-ing a better convergence state of the trained model, reduc-ing over-fitting on a given dataset." @default.
- W2892635580 created "2018-10-05" @default.
- W2892635580 creator A5013109703 @default.
- W2892635580 creator A5043783779 @default.
- W2892635580 creator A5078257438 @default.
- W2892635580 date "2018-09-25" @default.
- W2892635580 modified "2023-09-26" @default.
- W2892635580 title "Confidence Inference for Focused Learning in Stereo Matching." @default.
- W2892635580 cites W1522301498 @default.
- W2892635580 cites W1915985751 @default.
- W2892635580 cites W1939746761 @default.
- W2892635580 cites W1977192841 @default.
- W2892635580 cites W2105192278 @default.
- W2892635580 cites W2289083164 @default.
- W2892635580 cites W247920590 @default.
- W2892635580 cites W2752379698 @default.
- W2892635580 cites W2786643646 @default.
- W2892635580 cites W2794981464 @default.
- W2892635580 cites W2950094539 @default.
- W2892635580 cites W2950517871 @default.
- W2892635580 cites W63091017 @default.
- W2892635580 hasPublicationYear "2018" @default.
- W2892635580 type Work @default.
- W2892635580 sameAs 2892635580 @default.
- W2892635580 citedByCount "0" @default.
- W2892635580 crossrefType "proceedings-article" @default.
- W2892635580 hasAuthorship W2892635580A5013109703 @default.
- W2892635580 hasAuthorship W2892635580A5043783779 @default.
- W2892635580 hasAuthorship W2892635580A5078257438 @default.
- W2892635580 hasConcept C105795698 @default.
- W2892635580 hasConcept C119857082 @default.
- W2892635580 hasConcept C121158502 @default.
- W2892635580 hasConcept C153180895 @default.
- W2892635580 hasConcept C154945302 @default.
- W2892635580 hasConcept C160633673 @default.
- W2892635580 hasConcept C161191863 @default.
- W2892635580 hasConcept C165064840 @default.
- W2892635580 hasConcept C2776214188 @default.
- W2892635580 hasConcept C33923547 @default.
- W2892635580 hasConcept C41008148 @default.
- W2892635580 hasConcept C49937458 @default.
- W2892635580 hasConceptScore W2892635580C105795698 @default.
- W2892635580 hasConceptScore W2892635580C119857082 @default.
- W2892635580 hasConceptScore W2892635580C121158502 @default.
- W2892635580 hasConceptScore W2892635580C153180895 @default.
- W2892635580 hasConceptScore W2892635580C154945302 @default.
- W2892635580 hasConceptScore W2892635580C160633673 @default.
- W2892635580 hasConceptScore W2892635580C161191863 @default.
- W2892635580 hasConceptScore W2892635580C165064840 @default.
- W2892635580 hasConceptScore W2892635580C2776214188 @default.
- W2892635580 hasConceptScore W2892635580C33923547 @default.
- W2892635580 hasConceptScore W2892635580C41008148 @default.
- W2892635580 hasConceptScore W2892635580C49937458 @default.
- W2892635580 hasLocation W28926355801 @default.
- W2892635580 hasOpenAccess W2892635580 @default.
- W2892635580 hasPrimaryLocation W28926355801 @default.
- W2892635580 hasRelatedWork W150235323 @default.
- W2892635580 hasRelatedWork W1547922449 @default.
- W2892635580 hasRelatedWork W1584051547 @default.
- W2892635580 hasRelatedWork W2101496309 @default.
- W2892635580 hasRelatedWork W2289083164 @default.
- W2892635580 hasRelatedWork W2739299403 @default.
- W2892635580 hasRelatedWork W2739489527 @default.
- W2892635580 hasRelatedWork W2752144486 @default.
- W2892635580 hasRelatedWork W2912099768 @default.
- W2892635580 hasRelatedWork W2948953016 @default.
- W2892635580 hasRelatedWork W2997498246 @default.
- W2892635580 hasRelatedWork W3004672782 @default.
- W2892635580 hasRelatedWork W3016892155 @default.
- W2892635580 hasRelatedWork W3045919984 @default.
- W2892635580 hasRelatedWork W3047787625 @default.
- W2892635580 hasRelatedWork W3161721979 @default.
- W2892635580 hasRelatedWork W3161740746 @default.
- W2892635580 hasRelatedWork W3162192368 @default.
- W2892635580 hasRelatedWork W3189098873 @default.
- W2892635580 hasRelatedWork W2245561616 @default.
- W2892635580 isParatext "false" @default.
- W2892635580 isRetracted "false" @default.
- W2892635580 magId "2892635580" @default.
- W2892635580 workType "article" @default.