Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385482810> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4385482810 abstract "In deep learning, unsupervised domain adaptation (UDA) is commonly utilized when the availability of abundant labeled data is often limited. Several methods have been proposed for UDA to overcome the difficulty of distinguishing between semantically similar classes, such as person vs. rider and road vs. sidewalk. The confusion of the classes results from the collapse of the distance, caused by the domain shift, between classes in the feature space. In this work, we present a versatile approach based on text-image correlation-guided domain adaptation (TigDA), which maintains a distance to properly adjust the decision boundaries between classes in the feature space. In our approach, the feature information is extracted through text embedding of classes and the aligning capability of the text features with the image features is achieved using the cross-modality. The resultant cross-modal features play an essential role in generating pseudo-labels and calculating an auxiliary pixel-wise cross-entropy loss to assist the image encoder in learning the distribution of cross-modal features. Such a guiding process allows the extension of the distance between similar classes in feature space so that a proper distance for adjusting the decision boundary is maintained. Our TigDA achieved the highest performance among other UDA methods in both single-resolution and multi-resolution cases with the help of GTA5 and SYNTHIA for the source domain and Cityscapes for the target domain. The simplicity and versatility of TigDA will be widely applicable for enhancing the self-training capabilities of most UDA methods." @default.
- W4385482810 created "2023-08-03" @default.
- W4385482810 creator A5009191669 @default.
- W4385482810 creator A5011014617 @default.
- W4385482810 creator A5018682136 @default.
- W4385482810 creator A5068214919 @default.
- W4385482810 date "2023-06-18" @default.
- W4385482810 modified "2023-09-26" @default.
- W4385482810 title "Enhancing Discriminative Ability among Similar Classes with Guidance of Text-Image Correlation for Unsupervised Domain Adaptation" @default.
- W4385482810 cites W1978060137 @default.
- W4385482810 cites W2061247098 @default.
- W4385482810 cites W2194775991 @default.
- W4385482810 cites W2340897893 @default.
- W4385482810 cites W2431874326 @default.
- W4385482810 cites W2593768305 @default.
- W4385482810 cites W2616866739 @default.
- W4385482810 cites W2963118547 @default.
- W4385482810 cites W3120804725 @default.
- W4385482810 cites W3175294391 @default.
- W4385482810 cites W3198377975 @default.
- W4385482810 cites W3200888063 @default.
- W4385482810 cites W3202422321 @default.
- W4385482810 cites W3216551675 @default.
- W4385482810 cites W3217147624 @default.
- W4385482810 cites W4225649887 @default.
- W4385482810 cites W4226058394 @default.
- W4385482810 cites W4226396592 @default.
- W4385482810 cites W4287124998 @default.
- W4385482810 cites W4312310776 @default.
- W4385482810 cites W4312420092 @default.
- W4385482810 cites W4312599396 @default.
- W4385482810 cites W4312768455 @default.
- W4385482810 cites W4312821800 @default.
- W4385482810 cites W4312960937 @default.
- W4385482810 doi "https://doi.org/10.1109/ijcnn54540.2023.10191689" @default.
- W4385482810 hasPublicationYear "2023" @default.
- W4385482810 type Work @default.
- W4385482810 citedByCount "0" @default.
- W4385482810 crossrefType "proceedings-article" @default.
- W4385482810 hasAuthorship W4385482810A5009191669 @default.
- W4385482810 hasAuthorship W4385482810A5011014617 @default.
- W4385482810 hasAuthorship W4385482810A5018682136 @default.
- W4385482810 hasAuthorship W4385482810A5068214919 @default.
- W4385482810 hasConcept C106301342 @default.
- W4385482810 hasConcept C111919701 @default.
- W4385482810 hasConcept C115961682 @default.
- W4385482810 hasConcept C118505674 @default.
- W4385482810 hasConcept C121332964 @default.
- W4385482810 hasConcept C138885662 @default.
- W4385482810 hasConcept C153180895 @default.
- W4385482810 hasConcept C154945302 @default.
- W4385482810 hasConcept C2776401178 @default.
- W4385482810 hasConcept C41008148 @default.
- W4385482810 hasConcept C41608201 @default.
- W4385482810 hasConcept C41895202 @default.
- W4385482810 hasConcept C62520636 @default.
- W4385482810 hasConcept C75294576 @default.
- W4385482810 hasConcept C83665646 @default.
- W4385482810 hasConcept C97931131 @default.
- W4385482810 hasConceptScore W4385482810C106301342 @default.
- W4385482810 hasConceptScore W4385482810C111919701 @default.
- W4385482810 hasConceptScore W4385482810C115961682 @default.
- W4385482810 hasConceptScore W4385482810C118505674 @default.
- W4385482810 hasConceptScore W4385482810C121332964 @default.
- W4385482810 hasConceptScore W4385482810C138885662 @default.
- W4385482810 hasConceptScore W4385482810C153180895 @default.
- W4385482810 hasConceptScore W4385482810C154945302 @default.
- W4385482810 hasConceptScore W4385482810C2776401178 @default.
- W4385482810 hasConceptScore W4385482810C41008148 @default.
- W4385482810 hasConceptScore W4385482810C41608201 @default.
- W4385482810 hasConceptScore W4385482810C41895202 @default.
- W4385482810 hasConceptScore W4385482810C62520636 @default.
- W4385482810 hasConceptScore W4385482810C75294576 @default.
- W4385482810 hasConceptScore W4385482810C83665646 @default.
- W4385482810 hasConceptScore W4385482810C97931131 @default.
- W4385482810 hasFunder F4320324891 @default.
- W4385482810 hasLocation W43854828101 @default.
- W4385482810 hasOpenAccess W4385482810 @default.
- W4385482810 hasPrimaryLocation W43854828101 @default.
- W4385482810 hasRelatedWork W1461621550 @default.
- W4385482810 hasRelatedWork W2020350089 @default.
- W4385482810 hasRelatedWork W2038828201 @default.
- W4385482810 hasRelatedWork W2518873948 @default.
- W4385482810 hasRelatedWork W2773500201 @default.
- W4385482810 hasRelatedWork W2905846897 @default.
- W4385482810 hasRelatedWork W2944661354 @default.
- W4385482810 hasRelatedWork W2970216048 @default.
- W4385482810 hasRelatedWork W3119216324 @default.
- W4385482810 hasRelatedWork W2990896888 @default.
- W4385482810 isParatext "false" @default.
- W4385482810 isRetracted "false" @default.
- W4385482810 workType "article" @default.