mlsanityData to 7 Oct 2026

Delving Deep into Pixel Alignment Feature for Accurate Multi-view Human Mesh Recovery

arXiv 2301.06020 ↗First posted 15 Jan 2023cs.CV eess.IVcited by 10

Not on the map. Why?

Abstract (arXiv)

Regression-based methods have shown high efficiency and effectiveness for multi-view human mesh recovery. The key components of a typical regressor lie in the feature extraction of input views and the fusion of multi-view features. In this paper, we present Pixel-aligned Feedback Fusion (PaFF) for accurate yet efficient human mesh recovery from multi-view images. PaFF is an iterative regression framework that performs feature extraction and fusion alternately. At each iteration, PaFF extracts pixel-aligned feedback features from each input view according to the reprojection of the current estimation and fuses them together with respect to each vertex of the downsampled mesh. In this way, our regressor can not only perceive the misalignment status of each view from the feedback features but also correct the mesh parameters more effectively based on the feature fusion on mesh vertices. Additionally, our regressor disentangles the global orientation and translation of the body mesh from the estimation of mesh parameters such that the camera parameters of input views can be better utilized in the regression process. The efficacy of our method is validated in the Human3.6M dataset via comprehensive ablation experiments, where PaFF achieves 33.02 MPJPE and brings significant improvements over the previous best solutions by more than 29%. The project page with code and video results can be found at https://kairobo.github.io/PaFF/.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 50 of its 63 references that are in the map. First posted on arXiv 15 Jan 2023.

Builds on

50 of 63 references.

References · 50

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  5. 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,186
  6. 2017End-to-end Recovery of Human Shape and Posecited by 883
  7. 2016Keep it SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Imagecited by 692
  8. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 666
  9. 2019Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loopcited by 625
  10. 2019VIBE: Video Inference for Human Body Pose and Shape Estimationcited by 615
  11. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 493
  12. 2022TEMOS: Generating diverse human motions from textual descriptionscited by 486
38 more
  1. 2020I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Imagecited by 274
  2. 2019Convolutional Mesh Regression for Single-Image Human Shape Reconstructioncited by 270
  3. 2021Mesh Graphormercited by 264
  4. 2019Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inferencecited by 252
  5. 2021PyMAF: 3D Human Pose and Shape Regression with Pyramidal Mesh Alignment Feedback Loopcited by 250
  6. 2020PaMIR: Parametric Model-Conditioned Implicit Representation for Image-based Human Reconstructioncited by 231
  7. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 220
  8. 2020Monocular, One-stage, Regression of Multiple 3D Peoplecited by 216
  9. 2018Neural Body Fitting: Unifying Deep Learning and Model-Based Human Pose and Shape Estimationcited by 185
  10. 2021FrankMocap: A Monocular 3D Whole-Body Pose Estimation System via Regression and Integrationcited by 168
  11. 2018Monocular Total Capture: Posing Face, Body, and Hands in the Wildcited by 162
  12. 2019Learnable Triangulation of Human Posecited by 160
  13. 2021Probabilistic Modeling for Human Mesh Recoverycited by 152
  14. 2022PyMAF-X: Towards Well-aligned Full-body Model Regression from Monocular Imagescited by 152
  15. 2018BodyNet: Volumetric Inference of 3D Human Body Shapescited by 141
  16. 2015Human Pose Estimation with Iterative Error Feedbackcited by 121
  17. 2020Monocular Real-time Hand Shape and Motion Capture using Multi-modal Datacited by 106
  18. 2022Recovering 3D Human Mesh from Monocular Images: A Surveycited by 103
  19. 2017Towards Accurate Markerless Human Shape and Pose Estimation over Timecited by 92
  20. 2022Interacting Attention Graph for Single Image Two-Hand Reconstructioncited by 87
  21. 2019How do neural networks see depth in single images?cited by 82
  22. 2020Human Body Model Fitting by Learned Gradient Descentcited by 81
  23. 2021Direct Multi-view Multi-person 3D Pose Estimationcited by 64
  24. 2020Neural Descent for Visual 3D Human Pose and Shapecited by 60
  25. 20204D Association Graph for Realtime Multi-person Motion Capture Using Multiple Video Camerascited by 49
  26. 2019Learning 3D Human Shape and Pose from Dense Body Partscited by 48
  27. 2021Probabilistic 3D Human Shape and Pose Estimation from Multiple Unconstrained Images in the Wildcited by 40
  28. 2019Delving Deep Into Hybrid Annotations for 3D Human Recovery in the Wildcited by 38
  29. 2021Graph-Based 3D Multi-Person Pose Estimation Using Multi-View Imagescited by 38
  30. 2022Learned Vertex Descent: A New Direction for 3D Human Model Fittingcited by 37
  31. 2021Lightweight Multi-person Total Motion Capture Using Sparse Multi-view Camerascited by 32
  32. 2019Shape-Aware Human Pose and Shape Reconstruction Using Multi-View Imagescited by 28
  33. 2021Dynamic Multi-Person Mesh Recovery From Uncalibrated Multi-View Camerascited by 26
  34. 2021LegoFormer: Transformers for Block-by-Block Multi-view 3D Reconstructioncited by 25
  35. 2016FlyCap: Markerless Motion Capture Using Multiple Autonomous Flying Camerascited by 23
  36. 2021Adaptive Multi-view and Temporal Fusing Transformer for 3D Human Pose Estimationcited by 19
  37. 2020Multi-view Human Pose and Shape Estimation Using Learnable Volumetric Aggregationcited by 10
  38. 2020Multi-View Consistency Loss for Improved Single-Image 3D Reconstruction of Clothed Peoplecited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 6; 2025: 1; 2026: 3.

10 papers.

  1. 2024MUC: Mixture of Uncalibrated Cameras for Robust 3D Human Body Reconstructioncited by 5
  2. 2024Human Mesh Recovery from Arbitrary Multi-view Imagescited by 5
  3. 2024CondiMen: Conditional Multi-Person Mesh Recoverycited by 5
  4. 2024SkelFormer: Markerless 3D Pose and Shape Estimation using Skeletal Transformerscited by 2
  5. 2024HeatFormer: A Neural Optimizer for Multiview Human Mesh Recoverycited by 2
  6. 2024AdaptiveFusion: Adaptive Multi-Modal Multi-View Fusion for 3D Human Body Reconstructioncited by 1
  7. 2026DiffProxy: Multi-View Human Mesh Recovery via Diffusion-Generated Dense Proxiescited by 1
  8. 2026Unconstrained Multi-view Human Pose Estimation with Algebraic Priorscited by 1
  9. 2025On the Use of Hierarchical Vision Foundation Models for Low-Cost Human Mesh Recovery and Pose Estimationcited by 0
  10. 2026Monocular Models are Strong Learners for Multi-View Human Mesh Recoverycited by 0