HumanMeshNet: Polygonal Mesh Recovery of Humans

Venkat, Abbhinav; Patel, Chaitanya; Agrawal, Yudhik; Sharma, Avinash

doi:10.1109/iccvw.2019.00273

Cited by 15 publications

(5 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods for modelling details like hair/cloth/skin by estimating offsets from SMPL vertex have been proposed, but they work on very tight clothing and can not model the loose clothing deformation arising from pose . (Bhatnagar et al, 2019;Venkat et al, 2019;Kolotouros et al, 2019b).…”

Section: Related Workmentioning

confidence: 99%

SHARP: Shape-Aware Reconstruction of People in Loose Clothing

Jinka¹,

Chacko²,

Srivastava³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recent advancements in deep learning have enabled 3D human body reconstruction from a monocular image, which has broad applications in multiple domains. In this paper, we propose SHARP (SHape Aware Reconstruction of People in loose clothing), a novel end-to-end trainable network that accurately recovers the 3D geometry and appearance of humans in loose clothing from a monocular image. SHARP uses a sparse and efficient fusion strategy to combine parametric body prior with a non-parametric 2D representation of clothed humans. The parametric body prior enforces geometrical consistency on the body shape and pose, while the non-parametric representation models loose clothing and handles self-occlusions as well. We also leverage the sparseness of the non-parametric representation for faster training of our network while using losses on 2D maps. Another key contribution is 3DHumans, our new life-like dataset of 3D human body scans with rich geometrical and textural details. We evaluate SHARP on 3DHumans and other publicly available datasets, and show superior qualitative and quantitative performance than existing state-of-the-art methods.

show abstract

Section: Related Workmentioning

confidence: 99%

SHARP: Shape-Aware Reconstruction of People in Loose Clothing

Jinka¹,

Chacko²,

Srivastava³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…More recently, instead of predicting parameters of the model, model-free methods that directly regress each vertex were proposed to avoid representation issues [30,37]. Venkat et al [50] captured the local deformation by learning the "implicitly structure" of human mesh. Similarly, Kolotouros et al [22] directly regressed the vertices of a template mesh to explore the topological structure explicitly.…”

Section: D Human Pose and Shape Estimationmentioning

confidence: 99%

“…Hugo et al [13] proposed to generate 3D clothed human with graph convolution variational Auto-Encoder (GCVAE). Simultaneously, some works [10,50] intended to deal with mesh vertices as point clouds for capturing deformation. Similarly, we also adopt GCN to process the mesh structure.…”

Section: Graph Convolution Network For 3d Shape Reconstructionmentioning

confidence: 99%

DC-GNet: Deep Mesh Relation Capturing Graph Convolution Network for 3D Human Shape Reconstruction

Zhou

Jiang

Cai

et al. 2021

Proceedings of the 29th ACM International Conference on Multimedia

View full text Add to dashboard Cite

In this paper, we aim to reconstruct a full 3D human shape from a single image. Previous vertex-level and parameter regression approaches reconstruct 3D human shape based on a pre-defined adjacency matrix to encode positive relations between nodes. The deep topological relations for the surface of the 3D human body are not carefully exploited. Moreover, the performance of most existing approaches often suffer from domain gap when handling more occlusion cases in real-world scenes.In this work, we propose a Deep Mesh Relation Capturing Graph Convolution Network, DC-GNet, with a shape completion task for 3D human shape reconstruction. Firstly, we propose to capture deep relations within mesh vertices, where an adaptive matrix encoding both positive and negative relations is introduced. Secondly, we propose a shape completion task to learn prior about various kinds of occlusion cases. Our approach encodes mesh structure from more subtle relations between nodes in a more distant region. Furthermore, our shape completion module alleviates the performance degradation issue in the outdoor scene. Extensive experiments on several benchmarks show that our approach outperforms the previous 3D human pose and shape estimation approaches. CCS CONCEPTS• Computing methodologies → Computer vision.

show abstract

“…Registration plays an important role in shape estimation, tracking and 3D model building [LMR*15, LAGP09, SKR*15, CK15, BKL*16, BRPB17, VPAS19], which aligns 3D scans with each other or a template. Many methods [TCL*13] have been proposed in the last several decades, here we only discuss registration methods for the articulated objects like body and hands.…”

Section: Related Workmentioning

confidence: 99%