DC Field | Value | Language |
---|---|---|
dc.contributor.author | Shin, Inkyu | ko |
dc.contributor.author | Tsai, Yi-Hsuan | ko |
dc.contributor.author | Schulter, Samuel | ko |
dc.contributor.author | Zhuang, Bingbing | ko |
dc.contributor.author | Liu, Buyu | ko |
dc.contributor.author | Garg, Sparsh | ko |
dc.contributor.author | Kweon, In-So | ko |
dc.contributor.author | Yoon, Kuk-Jin | ko |
dc.date.accessioned | 2022-11-02T08:02:00Z | - |
dc.date.available | 2022-11-02T08:02:00Z | - |
dc.date.created | 2022-03-08 | - |
dc.date.created | 2022-03-08 | - |
dc.date.created | 2022-03-08 | - |
dc.date.issued | 2022-06-24 | - |
dc.identifier.citation | IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022, pp.16907 - 16916 | - |
dc.identifier.issn | 1063-6919 | - |
dc.identifier.uri | http://hdl.handle.net/10203/299286 | - |
dc.description.abstract | Test-time adaptation approaches have recently emerged as a practical solution for handling domain shift without access to the source domain data. In this paper, we propose and explore a new multi-modal extension of test-time adaptation for 3D semantic segmentation. We find that, directly applying existing methods usually results in performance instability at test time, because multi-modal input is not considered jointly. To design a framework that can take All advantage of multi-modality, where each modality provides regularized self-supervisory signals to other modalities, we propose two complementary modules within and across the modalities. First, Intra-modal Pseudo-label Generation (Infra-PG) is introduced to obtain reliable pseudo labels within each modality by aggregating information from two models that are both pre-trained on source data but updated with target data at different paces. Second, Inter-modal Pseudo-label Refinement (Inter-PR) adaptively selects more reliable pseudo labels from different modalities based on a proposed consistency scheme. Experiments demonstrate that our regularized pseudo labels produce stable self-learning signals in numerous multi-modal test-time adaptation scenarios for 3D semantic segmentation. Visit our project website at https: //www.nec-labs.com/mas/MM-TTA | - |
dc.language | English | - |
dc.publisher | Computer Vision Foundation, IEEE Computer Society | - |
dc.title | MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentation | - |
dc.type | Conference | - |
dc.identifier.wosid | 000870783002070 | - |
dc.identifier.scopusid | 2-s2.0-85138885506 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 16907 | - |
dc.citation.endingpage | 16916 | - |
dc.citation.publicationname | IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Virtual | - |
dc.identifier.doi | 10.1109/CVPR52688.2022.01642 | - |
dc.contributor.localauthor | Kweon, In-So | - |
dc.contributor.localauthor | Yoon, Kuk-Jin | - |
dc.contributor.nonIdAuthor | Tsai, Yi-Hsuan | - |
dc.contributor.nonIdAuthor | Schulter, Samuel | - |
dc.contributor.nonIdAuthor | Zhuang, Bingbing | - |
dc.contributor.nonIdAuthor | Liu, Buyu | - |
dc.contributor.nonIdAuthor | Garg, Sparsh | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.