ED<sup>2</sup>IF<sup>2</sup>-Net: Learning Disentangled Deformed Implicit Fields and Enhanced Displacement Fields from Single Images Using Pyramid Vision Transformer
Authors: Zhu, X., Yao, X., Zhang, J., Zhu, M., You, L., Yang, X., Zhang, J., Zhao, H. and Zeng, D.
Journal: Applied Sciences (Switzerland)
Volume: 13
Issue: 13
eISSN: 2076-3417
DOI: 10.3390/app13137577
Abstract:There has emerged substantial research in addressing single-view 3D reconstruction and the majority of the state-of-the-art implicit methods employ CNNs as the backbone network. On the other hand, transformers have shown remarkable performance in many vision tasks. However, it is still unknown whether transformers are suitable for single-view implicit 3D reconstruction. In this paper, we propose the first end-to-end single-view 3D reconstruction network based on the Pyramid Vision Transformer (PVT), called (Formula presented.) -Net, which disentangles the reconstruction of an implicit field into the reconstruction of topological structures and the recovery of surface details to achieve high-fidelity shape reconstruction. (Formula presented.) -Net uses a Pyramid Vision Transformer encoder to extract multi-scale hierarchical local features and a global vector of the input single image, which are fed into three separate decoders. A coarse shape decoder reconstructs a coarse implicit field based on the global vector, a deformation decoder iteratively refines the coarse implicit field using the pixel-aligned local features to obtain a deformed implicit field through multiple implicit field deformation blocks (IFDBs), and a surface detail decoder predicts an enhanced displacement field using the local features with hybrid attention modules (HAMs). The final output is a fusion of the deformed implicit field and the enhanced displacement field, with four loss terms applied to reconstruct the coarse implicit field, structure details through a novel deformation loss, overall shape after fusion, and surface details via a Laplacian loss. The quantitative results obtained from the ShapeNet dataset validate the exceptional performance of (Formula presented.) -Net. Notably, (Formula presented.) -Net-L stands out as the top-performing variant, exhibiting the highest mean IoU, CD, EMD, ECD-3D, and ECD-2D scores, reaching impressive values of 61.1, 7.26, 2.51, 6.08, and 1.84, respectively. The extensive experimental evaluations consistently demonstrate the state-of-the-art capabilities of (Formula presented.) -Net in terms of reconstructing topological structures and recovering surface details, all while maintaining competitive inference time.
https://eprints.bournemouth.ac.uk/38821/
Source: Scopus
ED<SUP>2</SUP>IF<SUP>2</SUP>-Net: Learning Disentangled Deformed Implicit Fields and Enhanced Displacement Fields from Single Images Using Pyramid Vision Transformer
Authors: Zhu, X., Yao, X., Zhang, J., Zhu, M., You, L., Yang, X., Zhang, J., Zhao, H. and Zeng, D.
Journal: APPLIED SCIENCES-BASEL
Volume: 13
Issue: 13
eISSN: 2076-3417
DOI: 10.3390/app13137577
https://eprints.bournemouth.ac.uk/38821/
Source: Web of Science (Lite)
ED2IF2-Net: Learning Disentangled Deformed Implicit Fields and Enhanced Displacement Fields from Single Images Using Pyramid Vision Transformer
Authors: Zhu, X., Yao, X., Zhang, J., Zhu, M., You, L., Yang, X., Zhang, J.J., Zhao, H. and Zeng, D.
Journal: Applied Sciences
Volume: 13
Issue: 13
ISSN: 2076-3417
Abstract:There has emerged substantial research in addressing single-view 3D reconstruction and the majority of the state-of-the-art implicit methods employ CNNs as the backbone network. On the other hand, transformers have shown remarkable performance in many vision tasks. However, it is still unknown whether transformers are suitable for single-view implicit 3D reconstruction. In this paper, we propose the first end-to-end single-view 3D reconstruction network based on the Pyramid Vision Transformer (PVT), called (Formula presented.) -Net, which disentangles the reconstruction of an implicit field into the reconstruction of topological structures and the recovery of surface details to achieve high-fidelity shape reconstruction. (Formula presented.) -Net uses a Pyramid Vision Transformer encoder to extract multi-scale hierarchical local features and a global vector of the input single image, which are fed into three separate decoders. A coarse shape decoder reconstructs a coarse implicit field based on the global vector, a deformation decoder iteratively refines the coarse implicit field using the pixel-aligned local features to obtain a deformed implicit field through multiple implicit field deformation blocks (IFDBs), and a surface detail decoder predicts an enhanced displacement field using the local features with hybrid attention modules (HAMs). The final output is a fusion of the deformed implicit field and the enhanced displacement field, with four loss terms applied to reconstruct the coarse implicit field, structure details through a novel deformation loss, overall shape after fusion, and surface details via a Laplacian loss. The quantitative results obtained from the ShapeNet dataset validate the exceptional performance of (Formula presented.) -Net. Notably, (Formula presented.) -Net-L stands out as the top-performing variant, exhibiting the highest mean IoU, CD, EMD, ECD-3D, and ECD-2D scores, reaching impressive values of 61.1, 7.26, 2.51, 6.08, and 1.84, respectively. The extensive experimental evaluations consistently demonstrate the state-of-the-art capabilities of (Formula presented.) -Net in terms of reconstructing topological structures and recovering surface details, all while maintaining competitive inference time.
https://eprints.bournemouth.ac.uk/38821/
Source: BURO EPrints