Matches in SemOpenAlex for { <https://semopenalex.org/work/W3043276225> ?p ?o ?g. }
- W3043276225 abstract "Most state-of-the-art approaches to road extraction from aerial images rely on a CNN trained to label road pixels as foreground and remainder of the image as background. The CNN is usually trained by minimizing pixel-wise losses, which is less than ideal to produce binary masks that preserve the road network's global connectivity. To address this issue, we introduce an Adversarial Learning (AL) strategy tailored for our purposes. A naive one would treat the segmentation network as a generator and would feed its output along with ground-truth segmentations to a discriminator. It would then train the generator and discriminator jointly. We will show that this is not enough because it does not capture the fact that most errors are local and need to be treated as such. Instead, we use a more sophisticated discriminator that returns a label pyramid describing what portions of the road network are correct at several different scales. This discriminator and the structured labels it returns are what gives our approach its edge and we will show that it outperforms state-of-the-art ones on the challenging RoadTracer dataset." @default.
- W3043276225 created "2020-07-23" @default.
- W3043276225 creator A5006418598 @default.
- W3043276225 creator A5013073642 @default.
- W3043276225 creator A5038674741 @default.
- W3043276225 creator A5044530513 @default.
- W3043276225 date "2020-07-17" @default.
- W3043276225 modified "2023-10-01" @default.
- W3043276225 title "TopoAL: An Adversarial Learning Approach for Topology-Aware Road Segmentation" @default.
- W3043276225 cites W1901129140 @default.
- W3043276225 cites W1981329364 @default.
- W3043276225 cites W1987365705 @default.
- W3043276225 cites W2033815587 @default.
- W3043276225 cites W2060095247 @default.
- W3043276225 cites W2099471712 @default.
- W3043276225 cites W2142015647 @default.
- W3043276225 cites W2155226776 @default.
- W3043276225 cites W2242818861 @default.
- W3043276225 cites W2252268321 @default.
- W3043276225 cites W2593886839 @default.
- W3043276225 cites W2738588019 @default.
- W3043276225 cites W2774327564 @default.
- W3043276225 cites W2780861787 @default.
- W3043276225 cites W2798268958 @default.
- W3043276225 cites W2798925380 @default.
- W3043276225 cites W2804199516 @default.
- W3043276225 cites W2811199523 @default.
- W3043276225 cites W2893801697 @default.
- W3043276225 cites W2899771611 @default.
- W3043276225 cites W2945957599 @default.
- W3043276225 cites W2962976523 @default.
- W3043276225 cites W2963073614 @default.
- W3043276225 cites W2963270775 @default.
- W3043276225 cites W2963800363 @default.
- W3043276225 cites W2964121744 @default.
- W3043276225 cites W2975194617 @default.
- W3043276225 cites W2981989409 @default.
- W3043276225 cites W2982182858 @default.
- W3043276225 cites W2988499426 @default.
- W3043276225 cites W3022194838 @default.
- W3043276225 cites W73112891 @default.
- W3043276225 cites W2122166615 @default.
- W3043276225 hasPublicationYear "2020" @default.
- W3043276225 type Work @default.
- W3043276225 sameAs 3043276225 @default.
- W3043276225 citedByCount "1" @default.
- W3043276225 countsByYear W30432762252020 @default.
- W3043276225 crossrefType "posted-content" @default.
- W3043276225 hasAuthorship W3043276225A5006418598 @default.
- W3043276225 hasAuthorship W3043276225A5013073642 @default.
- W3043276225 hasAuthorship W3043276225A5038674741 @default.
- W3043276225 hasAuthorship W3043276225A5044530513 @default.
- W3043276225 hasConcept C111472728 @default.
- W3043276225 hasConcept C11413529 @default.
- W3043276225 hasConcept C115961682 @default.
- W3043276225 hasConcept C121332964 @default.
- W3043276225 hasConcept C138885662 @default.
- W3043276225 hasConcept C142575187 @default.
- W3043276225 hasConcept C146849305 @default.
- W3043276225 hasConcept C153180895 @default.
- W3043276225 hasConcept C154945302 @default.
- W3043276225 hasConcept C160633673 @default.
- W3043276225 hasConcept C162307627 @default.
- W3043276225 hasConcept C163258240 @default.
- W3043276225 hasConcept C2524010 @default.
- W3043276225 hasConcept C2776639384 @default.
- W3043276225 hasConcept C2779803651 @default.
- W3043276225 hasConcept C2780992000 @default.
- W3043276225 hasConcept C31972630 @default.
- W3043276225 hasConcept C33923547 @default.
- W3043276225 hasConcept C37736160 @default.
- W3043276225 hasConcept C41008148 @default.
- W3043276225 hasConcept C48103436 @default.
- W3043276225 hasConcept C48372109 @default.
- W3043276225 hasConcept C62520636 @default.
- W3043276225 hasConcept C76155785 @default.
- W3043276225 hasConcept C89600930 @default.
- W3043276225 hasConcept C94375191 @default.
- W3043276225 hasConcept C94915269 @default.
- W3043276225 hasConceptScore W3043276225C111472728 @default.
- W3043276225 hasConceptScore W3043276225C11413529 @default.
- W3043276225 hasConceptScore W3043276225C115961682 @default.
- W3043276225 hasConceptScore W3043276225C121332964 @default.
- W3043276225 hasConceptScore W3043276225C138885662 @default.
- W3043276225 hasConceptScore W3043276225C142575187 @default.
- W3043276225 hasConceptScore W3043276225C146849305 @default.
- W3043276225 hasConceptScore W3043276225C153180895 @default.
- W3043276225 hasConceptScore W3043276225C154945302 @default.
- W3043276225 hasConceptScore W3043276225C160633673 @default.
- W3043276225 hasConceptScore W3043276225C162307627 @default.
- W3043276225 hasConceptScore W3043276225C163258240 @default.
- W3043276225 hasConceptScore W3043276225C2524010 @default.
- W3043276225 hasConceptScore W3043276225C2776639384 @default.
- W3043276225 hasConceptScore W3043276225C2779803651 @default.
- W3043276225 hasConceptScore W3043276225C2780992000 @default.
- W3043276225 hasConceptScore W3043276225C31972630 @default.
- W3043276225 hasConceptScore W3043276225C33923547 @default.
- W3043276225 hasConceptScore W3043276225C37736160 @default.
- W3043276225 hasConceptScore W3043276225C41008148 @default.
- W3043276225 hasConceptScore W3043276225C48103436 @default.