Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386044578> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4386044578 abstract "Anomaly segmentation plays a crucial role in identifying anomalous objects within images, which facilitates the detection of road anomalies for autonomous driving. Although existing methods have shown impressive results in anomaly segmentation using synthetic training data, the domain discrepancies between synthetic training data and real test data are often neglected. To address this issue, the Multi-Granularity Cross-Domain Alignment (MGCDA) framework is proposed for anomaly segmentation in complex driving environments. It uniquely combines a new Multi-source Domain Adversarial Training (MDAT) module and a novel Cross-domain Anomaly-aware Contrastive Learning (CACL) method to boost the generality of the model, seamlessly integrating multi-domain data at both scene and sample levels. Multi-source domain adversarial loss and a dynamic label smoothing strategy are integrated into the MDAT module to facilitate the acquisition of domain-invariant features at the scene level, through adversarial training across multiple stages. CACL aligns sample-level representations with contrastive loss on cross-domain data, which utilizes an anomaly-aware sampling strategy to efficiently sample hard samples and anchors. The proposed framework has decent properties of parameter-free during the inference stage and is compatible with other anomaly segmentation networks. Experimental conducted on Fishyscapes and RoadAnomaly datasets demonstrate that the proposed framework achieves state-of-the-art performance." @default.
- W4386044578 created "2023-08-22" @default.
- W4386044578 creator A5027341812 @default.
- W4386044578 creator A5056823945 @default.
- W4386044578 creator A5058640298 @default.
- W4386044578 creator A5058898461 @default.
- W4386044578 creator A5087345726 @default.
- W4386044578 date "2023-08-16" @default.
- W4386044578 modified "2023-10-16" @default.
- W4386044578 title "Improving Anomaly Segmentation with Multi-Granularity Cross-Domain Alignment" @default.
- W4386044578 doi "https://doi.org/10.48550/arxiv.2308.08696" @default.
- W4386044578 hasPublicationYear "2023" @default.
- W4386044578 type Work @default.
- W4386044578 citedByCount "0" @default.
- W4386044578 crossrefType "posted-content" @default.
- W4386044578 hasAuthorship W4386044578A5027341812 @default.
- W4386044578 hasAuthorship W4386044578A5056823945 @default.
- W4386044578 hasAuthorship W4386044578A5058640298 @default.
- W4386044578 hasAuthorship W4386044578A5058898461 @default.
- W4386044578 hasAuthorship W4386044578A5087345726 @default.
- W4386044578 hasBestOaLocation W43860445781 @default.
- W4386044578 hasConcept C111919701 @default.
- W4386044578 hasConcept C119857082 @default.
- W4386044578 hasConcept C121332964 @default.
- W4386044578 hasConcept C124101348 @default.
- W4386044578 hasConcept C12997251 @default.
- W4386044578 hasConcept C134306372 @default.
- W4386044578 hasConcept C153180895 @default.
- W4386044578 hasConcept C154945302 @default.
- W4386044578 hasConcept C177774035 @default.
- W4386044578 hasConcept C185592680 @default.
- W4386044578 hasConcept C198531522 @default.
- W4386044578 hasConcept C26873012 @default.
- W4386044578 hasConcept C2776214188 @default.
- W4386044578 hasConcept C31972630 @default.
- W4386044578 hasConcept C33923547 @default.
- W4386044578 hasConcept C36503486 @default.
- W4386044578 hasConcept C3770464 @default.
- W4386044578 hasConcept C37736160 @default.
- W4386044578 hasConcept C41008148 @default.
- W4386044578 hasConcept C43617362 @default.
- W4386044578 hasConcept C739882 @default.
- W4386044578 hasConcept C89600930 @default.
- W4386044578 hasConceptScore W4386044578C111919701 @default.
- W4386044578 hasConceptScore W4386044578C119857082 @default.
- W4386044578 hasConceptScore W4386044578C121332964 @default.
- W4386044578 hasConceptScore W4386044578C124101348 @default.
- W4386044578 hasConceptScore W4386044578C12997251 @default.
- W4386044578 hasConceptScore W4386044578C134306372 @default.
- W4386044578 hasConceptScore W4386044578C153180895 @default.
- W4386044578 hasConceptScore W4386044578C154945302 @default.
- W4386044578 hasConceptScore W4386044578C177774035 @default.
- W4386044578 hasConceptScore W4386044578C185592680 @default.
- W4386044578 hasConceptScore W4386044578C198531522 @default.
- W4386044578 hasConceptScore W4386044578C26873012 @default.
- W4386044578 hasConceptScore W4386044578C2776214188 @default.
- W4386044578 hasConceptScore W4386044578C31972630 @default.
- W4386044578 hasConceptScore W4386044578C33923547 @default.
- W4386044578 hasConceptScore W4386044578C36503486 @default.
- W4386044578 hasConceptScore W4386044578C3770464 @default.
- W4386044578 hasConceptScore W4386044578C37736160 @default.
- W4386044578 hasConceptScore W4386044578C41008148 @default.
- W4386044578 hasConceptScore W4386044578C43617362 @default.
- W4386044578 hasConceptScore W4386044578C739882 @default.
- W4386044578 hasConceptScore W4386044578C89600930 @default.
- W4386044578 hasLocation W43860445781 @default.
- W4386044578 hasOpenAccess W4386044578 @default.
- W4386044578 hasPrimaryLocation W43860445781 @default.
- W4386044578 hasRelatedWork W1976030966 @default.
- W4386044578 hasRelatedWork W2042251007 @default.
- W4386044578 hasRelatedWork W2065643612 @default.
- W4386044578 hasRelatedWork W2076520961 @default.
- W4386044578 hasRelatedWork W2110365568 @default.
- W4386044578 hasRelatedWork W2808998370 @default.
- W4386044578 hasRelatedWork W2974723675 @default.
- W4386044578 hasRelatedWork W2984111956 @default.
- W4386044578 hasRelatedWork W4323256105 @default.
- W4386044578 hasRelatedWork W4385644078 @default.
- W4386044578 isParatext "false" @default.
- W4386044578 isRetracted "false" @default.
- W4386044578 workType "article" @default.