Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387209096> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4387209096 abstract "We present a new pre-training strategy called M$^{3}$3D ($underline{M}$ulti-$underline{M}$odal $underline{M}$asked $underline{3D}$) built based on Multi-modal masked autoencoders that can leverage 3D priors and learned cross-modal representations in RGB-D data. We integrate two major self-supervised learning frameworks; Masked Image Modeling (MIM) and contrastive learning; aiming to effectively embed masked 3D priors and modality complementary features to enhance the correspondence between modalities. In contrast to recent approaches which are either focusing on specific downstream tasks or require multi-view correspondence, we show that our pre-training strategy is ubiquitous, enabling improved representation learning that can transfer into improved performance on various downstream tasks such as video action recognition, video action detection, 2D semantic segmentation and depth estimation. Experiments show that M$^{3}$3D outperforms the existing state-of-the-art approaches on ScanNet, NYUv2, UCF-101 and OR-AR, particularly with an improvement of +1.3% mIoU against Mask3D on ScanNet semantic segmentation. We further evaluate our method on low-data regime and demonstrate its superior data efficiency compared to current state-of-the-art approaches." @default.
- W4387209096 created "2023-09-30" @default.
- W4387209096 creator A5041390353 @default.
- W4387209096 creator A5067122011 @default.
- W4387209096 date "2023-09-26" @default.
- W4387209096 modified "2023-10-01" @default.
- W4387209096 title "M$^{3}$3D: Learning 3D priors using Multi-Modal Masked Autoencoders for 2D image and video understanding" @default.
- W4387209096 doi "https://doi.org/10.48550/arxiv.2309.15313" @default.
- W4387209096 hasPublicationYear "2023" @default.
- W4387209096 type Work @default.
- W4387209096 citedByCount "0" @default.
- W4387209096 crossrefType "posted-content" @default.
- W4387209096 hasAuthorship W4387209096A5041390353 @default.
- W4387209096 hasAuthorship W4387209096A5067122011 @default.
- W4387209096 hasBestOaLocation W43872090961 @default.
- W4387209096 hasConcept C107673813 @default.
- W4387209096 hasConcept C119857082 @default.
- W4387209096 hasConcept C150899416 @default.
- W4387209096 hasConcept C153083717 @default.
- W4387209096 hasConcept C153180895 @default.
- W4387209096 hasConcept C154945302 @default.
- W4387209096 hasConcept C17744445 @default.
- W4387209096 hasConcept C177769412 @default.
- W4387209096 hasConcept C185592680 @default.
- W4387209096 hasConcept C188027245 @default.
- W4387209096 hasConcept C199539241 @default.
- W4387209096 hasConcept C2776359362 @default.
- W4387209096 hasConcept C2776502983 @default.
- W4387209096 hasConcept C31972630 @default.
- W4387209096 hasConcept C41008148 @default.
- W4387209096 hasConcept C51632099 @default.
- W4387209096 hasConcept C59404180 @default.
- W4387209096 hasConcept C71139939 @default.
- W4387209096 hasConcept C89600930 @default.
- W4387209096 hasConcept C94625758 @default.
- W4387209096 hasConceptScore W4387209096C107673813 @default.
- W4387209096 hasConceptScore W4387209096C119857082 @default.
- W4387209096 hasConceptScore W4387209096C150899416 @default.
- W4387209096 hasConceptScore W4387209096C153083717 @default.
- W4387209096 hasConceptScore W4387209096C153180895 @default.
- W4387209096 hasConceptScore W4387209096C154945302 @default.
- W4387209096 hasConceptScore W4387209096C17744445 @default.
- W4387209096 hasConceptScore W4387209096C177769412 @default.
- W4387209096 hasConceptScore W4387209096C185592680 @default.
- W4387209096 hasConceptScore W4387209096C188027245 @default.
- W4387209096 hasConceptScore W4387209096C199539241 @default.
- W4387209096 hasConceptScore W4387209096C2776359362 @default.
- W4387209096 hasConceptScore W4387209096C2776502983 @default.
- W4387209096 hasConceptScore W4387209096C31972630 @default.
- W4387209096 hasConceptScore W4387209096C41008148 @default.
- W4387209096 hasConceptScore W4387209096C51632099 @default.
- W4387209096 hasConceptScore W4387209096C59404180 @default.
- W4387209096 hasConceptScore W4387209096C71139939 @default.
- W4387209096 hasConceptScore W4387209096C89600930 @default.
- W4387209096 hasConceptScore W4387209096C94625758 @default.
- W4387209096 hasLocation W43872090961 @default.
- W4387209096 hasOpenAccess W4387209096 @default.
- W4387209096 hasPrimaryLocation W43872090961 @default.
- W4387209096 hasRelatedWork W1669643531 @default.
- W4387209096 hasRelatedWork W1982826852 @default.
- W4387209096 hasRelatedWork W1987421842 @default.
- W4387209096 hasRelatedWork W2005437358 @default.
- W4387209096 hasRelatedWork W2008656436 @default.
- W4387209096 hasRelatedWork W2023558673 @default.
- W4387209096 hasRelatedWork W2110230079 @default.
- W4387209096 hasRelatedWork W2134924024 @default.
- W4387209096 hasRelatedWork W2517104666 @default.
- W4387209096 hasRelatedWork W2613186388 @default.
- W4387209096 isParatext "false" @default.
- W4387209096 isRetracted "false" @default.
- W4387209096 workType "article" @default.