Matches in SemOpenAlex for { <https://semopenalex.org/work/W3118680908> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3118680908 abstract "We deal with the problem of cross-task and cross-domain knowledge transfer in the realm of scene understanding for autonomous vehicles. We consider the scenario where supervision is available for a pair of tasks in a source domain while it is available for only one of the tasks in the target domain. Given that, the goal is to perform inference for the task in the target which is devoid of any training information. We argue that the only reported work in learning across tasks and domains (AT/DT) [26] faces the problem of domain shift between the source and target domains, hindering predictions on the target domain when the transfer of knowledge is learned on a statistically different yet related source domain. As a remedy, we develop a novel framework called ADA-AT/DT based on the adversarial training strategy to ensure that the domain-gaps are minimized for the common cross-domain supervised task. This, in effect, helps in realizing a domain-independent task-transfer function that eventually helps in performing improved inference in the target domain. We demonstrate that our proposed method significantly outperforms [26] by using models with 81% fewer trainable parameters. In addition, we perform experiments on a transformation mapping similar to U-Net to ensure maximum exploitation of features for task transfer. Extensive experiments have been performed on four different domains (Synthia, CityScapes, Carla, and KITTI) for two visual tasks (depth estimation and semantic segmentation) to confirm the superiority of our method." @default.
- W3118680908 created "2021-01-18" @default.
- W3118680908 creator A5008041465 @default.
- W3118680908 creator A5016405213 @default.
- W3118680908 creator A5020786167 @default.
- W3118680908 creator A5079337487 @default.
- W3118680908 date "2021-01-01" @default.
- W3118680908 modified "2023-09-22" @default.
- W3118680908 title "ADA-AT/DT: An Adversarial Approach for Cross-Domain and Cross-Task Knowledge Transfer" @default.
- W3118680908 cites W1901129140 @default.
- W3118680908 cites W1920962657 @default.
- W3118680908 cites W2150066425 @default.
- W3118680908 cites W2214409633 @default.
- W3118680908 cites W2340897893 @default.
- W3118680908 cites W2343077198 @default.
- W3118680908 cites W2584009249 @default.
- W3118680908 cites W2593768305 @default.
- W3118680908 cites W2909393256 @default.
- W3118680908 cites W2937261546 @default.
- W3118680908 cites W2962793481 @default.
- W3118680908 cites W2963073614 @default.
- W3118680908 cites W2963430933 @default.
- W3118680908 cites W2963677766 @default.
- W3118680908 cites W2963877604 @default.
- W3118680908 cites W2963898943 @default.
- W3118680908 cites W2964185501 @default.
- W3118680908 cites W2968136324 @default.
- W3118680908 cites W2989631250 @default.
- W3118680908 cites W2989929904 @default.
- W3118680908 doi "https://doi.org/10.1109/wacv48630.2021.00354" @default.
- W3118680908 hasPublicationYear "2021" @default.
- W3118680908 type Work @default.
- W3118680908 sameAs 3118680908 @default.
- W3118680908 citedByCount "4" @default.
- W3118680908 countsByYear W31186809082021 @default.
- W3118680908 countsByYear W31186809082022 @default.
- W3118680908 countsByYear W31186809082023 @default.
- W3118680908 crossrefType "proceedings-article" @default.
- W3118680908 hasAuthorship W3118680908A5008041465 @default.
- W3118680908 hasAuthorship W3118680908A5016405213 @default.
- W3118680908 hasAuthorship W3118680908A5020786167 @default.
- W3118680908 hasAuthorship W3118680908A5079337487 @default.
- W3118680908 hasConcept C104317684 @default.
- W3118680908 hasConcept C119857082 @default.
- W3118680908 hasConcept C134306372 @default.
- W3118680908 hasConcept C150899416 @default.
- W3118680908 hasConcept C154945302 @default.
- W3118680908 hasConcept C162324750 @default.
- W3118680908 hasConcept C175154964 @default.
- W3118680908 hasConcept C185592680 @default.
- W3118680908 hasConcept C187736073 @default.
- W3118680908 hasConcept C204241405 @default.
- W3118680908 hasConcept C207685749 @default.
- W3118680908 hasConcept C2776214188 @default.
- W3118680908 hasConcept C2776960227 @default.
- W3118680908 hasConcept C2780451532 @default.
- W3118680908 hasConcept C33923547 @default.
- W3118680908 hasConcept C36503486 @default.
- W3118680908 hasConcept C41008148 @default.
- W3118680908 hasConcept C55493867 @default.
- W3118680908 hasConcept C56739046 @default.
- W3118680908 hasConceptScore W3118680908C104317684 @default.
- W3118680908 hasConceptScore W3118680908C119857082 @default.
- W3118680908 hasConceptScore W3118680908C134306372 @default.
- W3118680908 hasConceptScore W3118680908C150899416 @default.
- W3118680908 hasConceptScore W3118680908C154945302 @default.
- W3118680908 hasConceptScore W3118680908C162324750 @default.
- W3118680908 hasConceptScore W3118680908C175154964 @default.
- W3118680908 hasConceptScore W3118680908C185592680 @default.
- W3118680908 hasConceptScore W3118680908C187736073 @default.
- W3118680908 hasConceptScore W3118680908C204241405 @default.
- W3118680908 hasConceptScore W3118680908C207685749 @default.
- W3118680908 hasConceptScore W3118680908C2776214188 @default.
- W3118680908 hasConceptScore W3118680908C2776960227 @default.
- W3118680908 hasConceptScore W3118680908C2780451532 @default.
- W3118680908 hasConceptScore W3118680908C33923547 @default.
- W3118680908 hasConceptScore W3118680908C36503486 @default.
- W3118680908 hasConceptScore W3118680908C41008148 @default.
- W3118680908 hasConceptScore W3118680908C55493867 @default.
- W3118680908 hasConceptScore W3118680908C56739046 @default.
- W3118680908 hasLocation W31186809081 @default.
- W3118680908 hasOpenAccess W3118680908 @default.
- W3118680908 hasPrimaryLocation W31186809081 @default.
- W3118680908 hasRelatedWork W2964665600 @default.
- W3118680908 hasRelatedWork W2971007555 @default.
- W3118680908 hasRelatedWork W3118680908 @default.
- W3118680908 hasRelatedWork W3163380912 @default.
- W3118680908 hasRelatedWork W4200114345 @default.
- W3118680908 hasRelatedWork W4225633098 @default.
- W3118680908 hasRelatedWork W4226361678 @default.
- W3118680908 hasRelatedWork W4283787367 @default.
- W3118680908 hasRelatedWork W4283798916 @default.
- W3118680908 hasRelatedWork W4296300999 @default.
- W3118680908 isParatext "false" @default.
- W3118680908 isRetracted "false" @default.
- W3118680908 magId "3118680908" @default.
- W3118680908 workType "article" @default.