Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385474100> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4385474100 abstract "In this study, we focus on the problem of 3D human mesh recovery from a single image under obscured conditions. Most state-of-the-art methods aim to improve 2D alignment technologies, such as spatial averaging and 2D joint sampling. However, they tend to neglect the crucial aspect of 3D alignment by improving 3D representations. Furthermore, recent methods struggle to separate the target human from occlusion or background in crowded scenes as they optimize the 3D space of target human with 3D joint coordinates as local supervision. To address these issues, a desirable method would involve a framework for fusing 2D and 3D features and a strategy for optimizing the 3D space globally. Therefore, this paper presents 3D JOint contrastive learning with TRansformers (JOTR) framework for handling occluded 3D human mesh recovery. Our method includes an encoder-decoder transformer architecture to fuse 2D and 3D representations for achieving 2D$&$3D aligned results in a coarse-to-fine manner and a novel 3D joint contrastive learning approach for adding explicitly global supervision for the 3D feature space. The contrastive learning approach includes two contrastive losses: joint-to-joint contrast for enhancing the similarity of semantically similar voxels (i.e., human joints), and joint-to-non-joint contrast for ensuring discrimination from others (e.g., occlusions and background). Qualitative and quantitative analyses demonstrate that our method outperforms state-of-the-art competitors on both occlusion-specific and standard benchmarks, significantly improving the reconstruction of occluded humans." @default.
- W4385474100 created "2023-08-02" @default.
- W4385474100 creator A5005421447 @default.
- W4385474100 creator A5011321102 @default.
- W4385474100 creator A5020316185 @default.
- W4385474100 creator A5022703708 @default.
- W4385474100 creator A5042761483 @default.
- W4385474100 creator A5064884455 @default.
- W4385474100 date "2023-07-30" @default.
- W4385474100 modified "2023-10-18" @default.
- W4385474100 title "JOTR: 3D Joint Contrastive Learning with Transformers for Occluded Human Mesh Recovery" @default.
- W4385474100 doi "https://doi.org/10.48550/arxiv.2307.16377" @default.
- W4385474100 hasPublicationYear "2023" @default.
- W4385474100 type Work @default.
- W4385474100 citedByCount "0" @default.
- W4385474100 crossrefType "posted-content" @default.
- W4385474100 hasAuthorship W4385474100A5005421447 @default.
- W4385474100 hasAuthorship W4385474100A5011321102 @default.
- W4385474100 hasAuthorship W4385474100A5020316185 @default.
- W4385474100 hasAuthorship W4385474100A5022703708 @default.
- W4385474100 hasAuthorship W4385474100A5042761483 @default.
- W4385474100 hasAuthorship W4385474100A5064884455 @default.
- W4385474100 hasBestOaLocation W43854741001 @default.
- W4385474100 hasConcept C109950114 @default.
- W4385474100 hasConcept C111919701 @default.
- W4385474100 hasConcept C118505674 @default.
- W4385474100 hasConcept C119599485 @default.
- W4385474100 hasConcept C127413603 @default.
- W4385474100 hasConcept C153180895 @default.
- W4385474100 hasConcept C154945302 @default.
- W4385474100 hasConcept C165801399 @default.
- W4385474100 hasConcept C170154142 @default.
- W4385474100 hasConcept C18555067 @default.
- W4385474100 hasConcept C31972630 @default.
- W4385474100 hasConcept C41008148 @default.
- W4385474100 hasConcept C54170458 @default.
- W4385474100 hasConcept C59404180 @default.
- W4385474100 hasConcept C66322947 @default.
- W4385474100 hasConcept C83665646 @default.
- W4385474100 hasConceptScore W4385474100C109950114 @default.
- W4385474100 hasConceptScore W4385474100C111919701 @default.
- W4385474100 hasConceptScore W4385474100C118505674 @default.
- W4385474100 hasConceptScore W4385474100C119599485 @default.
- W4385474100 hasConceptScore W4385474100C127413603 @default.
- W4385474100 hasConceptScore W4385474100C153180895 @default.
- W4385474100 hasConceptScore W4385474100C154945302 @default.
- W4385474100 hasConceptScore W4385474100C165801399 @default.
- W4385474100 hasConceptScore W4385474100C170154142 @default.
- W4385474100 hasConceptScore W4385474100C18555067 @default.
- W4385474100 hasConceptScore W4385474100C31972630 @default.
- W4385474100 hasConceptScore W4385474100C41008148 @default.
- W4385474100 hasConceptScore W4385474100C54170458 @default.
- W4385474100 hasConceptScore W4385474100C59404180 @default.
- W4385474100 hasConceptScore W4385474100C66322947 @default.
- W4385474100 hasConceptScore W4385474100C83665646 @default.
- W4385474100 hasLocation W43854741001 @default.
- W4385474100 hasOpenAccess W4385474100 @default.
- W4385474100 hasPrimaryLocation W43854741001 @default.
- W4385474100 hasRelatedWork W2144724818 @default.
- W4385474100 hasRelatedWork W2160739731 @default.
- W4385474100 hasRelatedWork W2517246325 @default.
- W4385474100 hasRelatedWork W2685530223 @default.
- W4385474100 hasRelatedWork W2769668212 @default.
- W4385474100 hasRelatedWork W2980953096 @default.
- W4385474100 hasRelatedWork W2990870497 @default.
- W4385474100 hasRelatedWork W3157004382 @default.
- W4385474100 hasRelatedWork W4206076898 @default.
- W4385474100 hasRelatedWork W2494681155 @default.
- W4385474100 isParatext "false" @default.
- W4385474100 isRetracted "false" @default.
- W4385474100 workType "article" @default.