FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3149107907> ?p ?o ?g. }

• W3149107907 abstract "Simulators can efficiently generate large amounts of labeled synthetic data with perfect supervision for hard-to-label tasks like semantic segmentation. However, they introduce a domain gap that severely hurts real-world performance. We propose to use self-supervised monocular depth estimation as a proxy task to bridge this gap and improve sim-to-real unsupervised domain adaptation (UDA). Our Geometric Unsupervised Domain Adaptation method (GUDA) learns a domain-invariant representation via a multi-task objective combining synthetic semantic supervision with real-world geometric constraints on videos. GUDA establishes a new state of the art in UDA for semantic segmentation on three benchmarks, outperforming methods that use domain adversarial learning, self-training, or other self-supervised proxy tasks. Furthermore, we show that our method scales well with the quality and quantity of synthetic data while also improving depth prediction." @default.
• W3149107907 created "2021-04-13" @default.
• W3149107907 creator A5037522398 @default.
• W3149107907 creator A5063755961 @default.
• W3149107907 creator A5068252474 @default.
• W3149107907 creator A5075018873 @default.
• W3149107907 date "2021-03-30" @default.
• W3149107907 modified "2023-10-18" @default.
• W3149107907 title "Geometric Unsupervised Domain Adaptation for Semantic Segmentation" @default.
• W3149107907 cites W1529410181 @default.
• W3149107907 cites W1565327149 @default.
• W3149107907 cites W1776042733 @default.
• W3149107907 cites W1905829557 @default.
• W3149107907 cites W2108598243 @default.
• W3149107907 cites W2115579991 @default.
• W3149107907 cites W2133665775 @default.
• W3149107907 cites W2150066425 @default.
• W3149107907 cites W2171740948 @default.
• W3149107907 cites W2194775991 @default.
• W3149107907 cites W2300779272 @default.
• W3149107907 cites W2308529009 @default.
• W3149107907 cites W2321533354 @default.
• W3149107907 cites W2340897893 @default.
• W3149107907 cites W2397830550 @default.
• W3149107907 cites W2406270520 @default.
• W3149107907 cites W2431874326 @default.
• W3149107907 cites W2474531669 @default.
• W3149107907 cites W2478454054 @default.
• W3149107907 cites W2487365028 @default.
• W3149107907 cites W2557406251 @default.
• W3149107907 cites W2562192638 @default.
• W3149107907 cites W2562637781 @default.
• W3149107907 cites W2579628011 @default.
• W3149107907 cites W2590953969 @default.
• W3149107907 cites W2609883120 @default.
• W3149107907 cites W2786808285 @default.
• W3149107907 cites W2887368306 @default.
• W3149107907 cites W2895168809 @default.
• W3149107907 cites W2895281799 @default.
• W3149107907 cites W2899771611 @default.
• W3149107907 cites W2908510526 @default.
• W3149107907 cites W2952865063 @default.
• W3149107907 cites W2958360136 @default.
• W3149107907 cites W2962687275 @default.
• W3149107907 cites W2962742544 @default.
• W3149107907 cites W2962759496 @default.
• W3149107907 cites W2962808524 @default.
• W3149107907 cites W2962891704 @default.
• W3149107907 cites W2962960377 @default.
• W3149107907 cites W2963052201 @default.
• W3149107907 cites W2963073217 @default.
• W3149107907 cites W2963285578 @default.
• W3149107907 cites W2963481481 @default.
• W3149107907 cites W2963557767 @default.
• W3149107907 cites W2963865469 @default.
• W3149107907 cites W2964057616 @default.
• W3149107907 cites W2964126011 @default.
• W3149107907 cites W2964139811 @default.
• W3149107907 cites W2964277612 @default.
• W3149107907 cites W2964968086 @default.
• W3149107907 cites W2975758481 @default.
• W3149107907 cites W2980823292 @default.
• W3149107907 cites W2981429991 @default.
• W3149107907 cites W2982102242 @default.
• W3149107907 cites W2985406498 @default.
• W3149107907 cites W2985409929 @default.
• W3149107907 cites W2995884594 @default.
• W3149107907 cites W2998666297 @default.
• W3149107907 cites W3034247804 @default.
• W3149107907 cites W3034604951 @default.
• W3149107907 cites W3034642839 @default.
• W3149107907 cites W3034679848 @default.
• W3149107907 cites W3035294798 @default.
• W3149107907 cites W3039883906 @default.
• W3149107907 cites W3102977943 @default.
• W3149107907 cites W3107389224 @default.
• W3149107907 cites W3107909383 @default.
• W3149107907 cites W3108566666 @default.
• W3149107907 cites W3126596634 @default.
• W3149107907 hasPublicationYear "2021" @default.
• W3149107907 type Work @default.
• W3149107907 sameAs 3149107907 @default.
• W3149107907 citedByCount "0" @default.
• W3149107907 crossrefType "posted-content" @default.
• W3149107907 hasAuthorship W3149107907A5037522398 @default.
• W3149107907 hasAuthorship W3149107907A5063755961 @default.
• W3149107907 hasAuthorship W3149107907A5068252474 @default.
• W3149107907 hasAuthorship W3149107907A5075018873 @default.
• W3149107907 hasConcept C115961682 @default.
• W3149107907 hasConcept C119857082 @default.
• W3149107907 hasConcept C134306372 @default.
• W3149107907 hasConcept C153180895 @default.
• W3149107907 hasConcept C154945302 @default.
• W3149107907 hasConcept C160920958 @default.
• W3149107907 hasConcept C162324750 @default.
• W3149107907 hasConcept C1667742 @default.
• W3149107907 hasConcept C187736073 @default.
• W3149107907 hasConcept C2776145971 @default.
• W3149107907 hasConcept C2776434776 @default.
• W3149107907 hasConcept C2780451532 @default.

• next