Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313127614> ?p ?o ?g. }
- W4313127614 abstract "Semantic segmentation provides scene understanding capability by performing pixel-wise classification of objects within an image. However, the sensitivity of such algorithms towards domain changes requires fine-tuning using an annotated dataset for each novel domain, which is expensive to construct and inefficient. We highlight that irrespective of the training dataset, structural properties of scenes remain the same hence domain sensitivity arises from training methodology. Thus, in this paper, we propose a domain alignment approach wherein multiple synthetic source domains are used to train an underlying segmentation network such that it performs consistently in unknown real target domains. Towards this end, we propose a pixel-wise supervised contrastive learning framework that enforces constraints in latent space resulting in features belonging to the same class being clustered closely and away from different classes. This approach allows for better capturing of global and local semantics while providing domain invariant properties. Our approach can be easily incorporated into prior semantic segmentation approaches without the significant computational overhead. We empirically demonstrate the efficacy of the proposed approach on GTAV → Cityscapes, GTAV+Synthia → Cityscapes, and GTAV+Synthia+Synscapes → Cityscapes scenarios and report state-of-the-art (SoTA) performance without requiring access to images from the target domain." @default.
- W4313127614 created "2023-01-06" @default.
- W4313127614 creator A5012837410 @default.
- W4313127614 creator A5028140589 @default.
- W4313127614 creator A5058505853 @default.
- W4313127614 date "2022-10-23" @default.
- W4313127614 modified "2023-09-29" @default.
- W4313127614 title "Multi-Source Domain Alignment for Domain Invariant Segmentation in Unknown Targets" @default.
- W4313127614 cites W1978920452 @default.
- W4313127614 cites W1981658663 @default.
- W4313127614 cites W2194775991 @default.
- W4313127614 cites W2340897893 @default.
- W4313127614 cites W2412782625 @default.
- W4313127614 cites W2431874326 @default.
- W4313127614 cites W2583503935 @default.
- W4313127614 cites W2781228439 @default.
- W4313127614 cites W2798991696 @default.
- W4313127614 cites W2883090707 @default.
- W4313127614 cites W2895281799 @default.
- W4313127614 cites W2963052201 @default.
- W4313127614 cites W2963073217 @default.
- W4313127614 cites W2963107255 @default.
- W4313127614 cites W2963120918 @default.
- W4313127614 cites W2963292632 @default.
- W4313127614 cites W2963506806 @default.
- W4313127614 cites W2964309882 @default.
- W4313127614 cites W2981429991 @default.
- W4313127614 cites W2985409929 @default.
- W4313127614 cites W2986831462 @default.
- W4313127614 cites W2998607115 @default.
- W4313127614 cites W3014641072 @default.
- W4313127614 cites W3034272105 @default.
- W4313127614 cites W3034417116 @default.
- W4313127614 cites W3035086574 @default.
- W4313127614 cites W3035294798 @default.
- W4313127614 cites W3107909383 @default.
- W4313127614 cites W3113328489 @default.
- W4313127614 cites W3120804725 @default.
- W4313127614 cites W3122412340 @default.
- W4313127614 cites W3172292202 @default.
- W4313127614 cites W3174407746 @default.
- W4313127614 cites W3175294391 @default.
- W4313127614 cites W3180439858 @default.
- W4313127614 cites W3202422321 @default.
- W4313127614 cites W3207649350 @default.
- W4313127614 cites W4285813042 @default.
- W4313127614 cites W4312841491 @default.
- W4313127614 doi "https://doi.org/10.1109/iros47612.2022.9981166" @default.
- W4313127614 hasPublicationYear "2022" @default.
- W4313127614 type Work @default.
- W4313127614 citedByCount "0" @default.
- W4313127614 crossrefType "proceedings-article" @default.
- W4313127614 hasAuthorship W4313127614A5012837410 @default.
- W4313127614 hasAuthorship W4313127614A5028140589 @default.
- W4313127614 hasAuthorship W4313127614A5058505853 @default.
- W4313127614 hasConcept C111919701 @default.
- W4313127614 hasConcept C119857082 @default.
- W4313127614 hasConcept C124504099 @default.
- W4313127614 hasConcept C134306372 @default.
- W4313127614 hasConcept C153180895 @default.
- W4313127614 hasConcept C154945302 @default.
- W4313127614 hasConcept C160633673 @default.
- W4313127614 hasConcept C184337299 @default.
- W4313127614 hasConcept C190470478 @default.
- W4313127614 hasConcept C199360897 @default.
- W4313127614 hasConcept C2779960059 @default.
- W4313127614 hasConcept C31972630 @default.
- W4313127614 hasConcept C33923547 @default.
- W4313127614 hasConcept C36503486 @default.
- W4313127614 hasConcept C37914503 @default.
- W4313127614 hasConcept C41008148 @default.
- W4313127614 hasConcept C89600930 @default.
- W4313127614 hasConceptScore W4313127614C111919701 @default.
- W4313127614 hasConceptScore W4313127614C119857082 @default.
- W4313127614 hasConceptScore W4313127614C124504099 @default.
- W4313127614 hasConceptScore W4313127614C134306372 @default.
- W4313127614 hasConceptScore W4313127614C153180895 @default.
- W4313127614 hasConceptScore W4313127614C154945302 @default.
- W4313127614 hasConceptScore W4313127614C160633673 @default.
- W4313127614 hasConceptScore W4313127614C184337299 @default.
- W4313127614 hasConceptScore W4313127614C190470478 @default.
- W4313127614 hasConceptScore W4313127614C199360897 @default.
- W4313127614 hasConceptScore W4313127614C2779960059 @default.
- W4313127614 hasConceptScore W4313127614C31972630 @default.
- W4313127614 hasConceptScore W4313127614C33923547 @default.
- W4313127614 hasConceptScore W4313127614C36503486 @default.
- W4313127614 hasConceptScore W4313127614C37914503 @default.
- W4313127614 hasConceptScore W4313127614C41008148 @default.
- W4313127614 hasConceptScore W4313127614C89600930 @default.
- W4313127614 hasLocation W43131276141 @default.
- W4313127614 hasOpenAccess W4313127614 @default.
- W4313127614 hasPrimaryLocation W43131276141 @default.
- W4313127614 hasRelatedWork W121273120 @default.
- W4313127614 hasRelatedWork W1669643531 @default.
- W4313127614 hasRelatedWork W2005437358 @default.
- W4313127614 hasRelatedWork W2008656436 @default.
- W4313127614 hasRelatedWork W2023558673 @default.
- W4313127614 hasRelatedWork W2134924024 @default.
- W4313127614 hasRelatedWork W2337415362 @default.
- W4313127614 hasRelatedWork W2517104666 @default.