Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206488440> ?p ?o ?g. }
- W3206488440 abstract "Unsupervised Domain Adaptation (UDA) for semantic segmentation has been favorably applied to real-world scenarios in which pixel-level labels are hard to be obtained. In most of the existing UDA methods, all target data are assumed to be introduced simultaneously. Yet, the data are usually presented sequentially in the real world. Moreover, Continual UDA, which deals with more practical scenarios with multiple target domains in the continual learning setting, has not been actively explored. In this light, we propose Continual UDA for semantic segmentation based on a newly designed Expanding Target-specific Memory (ETM) framework. Our novel ETM framework contains Target-specific Memory (TM) for each target domain to alleviate catastrophic forgetting. Furthermore, a proposed Double Hinge Adversarial (DHA) loss leads the network to produce better UDA performance overall. Our design of the TM and training objectives let the semantic segmentation network adapt to the current target domain while preserving the knowledge learned on previous target domains. The model with the proposed framework outperforms other state-of-the-art models in continual learning settings on standard benchmarks such as GTA5, SYNTHIA, CityScapes, IDD, and Cross-City datasets. The source code is available at https://github.com/joonh-kim/ETM." @default.
- W3206488440 created "2021-10-25" @default.
- W3206488440 creator A5023474199 @default.
- W3206488440 creator A5031148632 @default.
- W3206488440 creator A5051029758 @default.
- W3206488440 creator A5058287337 @default.
- W3206488440 date "2021-10-11" @default.
- W3206488440 modified "2023-09-27" @default.
- W3206488440 title "Continual Unsupervised Domain Adaptation for Semantic Segmentation." @default.
- W3206488440 cites W1560724230 @default.
- W3206488440 cites W1682403713 @default.
- W3206488440 cites W1731081199 @default.
- W3206488440 cites W1745334888 @default.
- W3206488440 cites W1901129140 @default.
- W3206488440 cites W1903029394 @default.
- W3206488440 cites W2005295545 @default.
- W3206488440 cites W2037979274 @default.
- W3206488440 cites W2060277733 @default.
- W3206488440 cites W2099471712 @default.
- W3206488440 cites W2194775991 @default.
- W3206488440 cites W2340897893 @default.
- W3206488440 cites W2412782625 @default.
- W3206488440 cites W2431874326 @default.
- W3206488440 cites W2473930607 @default.
- W3206488440 cites W2487365028 @default.
- W3206488440 cites W2559597482 @default.
- W3206488440 cites W2560647685 @default.
- W3206488440 cites W2562192638 @default.
- W3206488440 cites W2565639579 @default.
- W3206488440 cites W2593768305 @default.
- W3206488440 cites W2630837129 @default.
- W3206488440 cites W2756154119 @default.
- W3206488440 cites W2781228439 @default.
- W3206488440 cites W2799166040 @default.
- W3206488440 cites W2886934227 @default.
- W3206488440 cites W2901870313 @default.
- W3206488440 cites W2912410476 @default.
- W3206488440 cites W2963073217 @default.
- W3206488440 cites W2963107255 @default.
- W3206488440 cites W2963120918 @default.
- W3206488440 cites W2963150697 @default.
- W3206488440 cites W2963540014 @default.
- W3206488440 cites W2963813679 @default.
- W3206488440 cites W2963826694 @default.
- W3206488440 cites W2970092410 @default.
- W3206488440 cites W2981512393 @default.
- W3206488440 cites W2981925632 @default.
- W3206488440 cites W2991405316 @default.
- W3206488440 cites W3034417116 @default.
- W3206488440 cites W3034478348 @default.
- W3206488440 cites W3035564946 @default.
- W3206488440 cites W3037695135 @default.
- W3206488440 cites W3044575716 @default.
- W3206488440 cites W3105326531 @default.
- W3206488440 cites W2426267443 @default.
- W3206488440 hasPublicationYear "2021" @default.
- W3206488440 type Work @default.
- W3206488440 sameAs 3206488440 @default.
- W3206488440 citedByCount "0" @default.
- W3206488440 crossrefType "posted-content" @default.
- W3206488440 hasAuthorship W3206488440A5023474199 @default.
- W3206488440 hasAuthorship W3206488440A5031148632 @default.
- W3206488440 hasAuthorship W3206488440A5051029758 @default.
- W3206488440 hasAuthorship W3206488440A5058287337 @default.
- W3206488440 hasConcept C119857082 @default.
- W3206488440 hasConcept C120665830 @default.
- W3206488440 hasConcept C121332964 @default.
- W3206488440 hasConcept C134306372 @default.
- W3206488440 hasConcept C138885662 @default.
- W3206488440 hasConcept C139807058 @default.
- W3206488440 hasConcept C154945302 @default.
- W3206488440 hasConcept C177264268 @default.
- W3206488440 hasConcept C184337299 @default.
- W3206488440 hasConcept C199360897 @default.
- W3206488440 hasConcept C2776145971 @default.
- W3206488440 hasConcept C2776434776 @default.
- W3206488440 hasConcept C2776760102 @default.
- W3206488440 hasConcept C33923547 @default.
- W3206488440 hasConcept C36503486 @default.
- W3206488440 hasConcept C41008148 @default.
- W3206488440 hasConcept C41895202 @default.
- W3206488440 hasConcept C7149132 @default.
- W3206488440 hasConcept C89600930 @default.
- W3206488440 hasConcept C95623464 @default.
- W3206488440 hasConceptScore W3206488440C119857082 @default.
- W3206488440 hasConceptScore W3206488440C120665830 @default.
- W3206488440 hasConceptScore W3206488440C121332964 @default.
- W3206488440 hasConceptScore W3206488440C134306372 @default.
- W3206488440 hasConceptScore W3206488440C138885662 @default.
- W3206488440 hasConceptScore W3206488440C139807058 @default.
- W3206488440 hasConceptScore W3206488440C154945302 @default.
- W3206488440 hasConceptScore W3206488440C177264268 @default.
- W3206488440 hasConceptScore W3206488440C184337299 @default.
- W3206488440 hasConceptScore W3206488440C199360897 @default.
- W3206488440 hasConceptScore W3206488440C2776145971 @default.
- W3206488440 hasConceptScore W3206488440C2776434776 @default.
- W3206488440 hasConceptScore W3206488440C2776760102 @default.
- W3206488440 hasConceptScore W3206488440C33923547 @default.
- W3206488440 hasConceptScore W3206488440C36503486 @default.
- W3206488440 hasConceptScore W3206488440C41008148 @default.