mlsanityData to 14 Oct 2026

TriHorn-Net: A Model for Accurate Depth-Based 3D Hand Pose Estimation

arXiv 2206.07117 ↗First posted 14 Jun 2022cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

3D hand pose estimation methods have made significant progress recently. However, the estimation accuracy is often far from sufficient for specific real-world applications, and thus there is significant room for improvement. This paper proposes TriHorn-Net, a novel model that uses specific innovations to improve hand pose estimation accuracy on depth images. The first innovation is the decomposition of the 3D hand pose estimation into the estimation of 2D joint locations in the depth image space (UV), and the estimation of their corresponding depths aided by two complementary attention maps. This decomposition prevents depth estimation, which is a more difficult task, from interfering with the UV estimations at both the prediction and feature levels. The second innovation is PixDropout, which is, to the best of our knowledge, the first appearance-based data augmentation method for hand depth images. Experimental results demonstrate that the proposed model outperforms the state-of-the-art methods on three public benchmark datasets. Our implementation is available at https://github.com/mrezaei92/TriHorn-Net.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 54 of its 108 references that are in the map. First posted on arXiv 14 Jun 2022.

Builds on

54 of 108 references.

References · 54

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  5. 2016Identity Mappings in Deep Residual Networkscited by 1,965
  6. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,628
  7. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,420
  8. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,380
  9. 2017Random Erasing Data Augmentationcited by 1,239
  10. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 915
  11. 2018AutoAugment: Learning Augmentation Policies from Datacited by 730
  12. 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
38 more
  1. 2017Neural 3D Mesh Renderercited by 437
  2. 2020I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Imagecited by 274
  3. 2017Integral Human Pose Regressioncited by 256
  4. 2017Learning to Estimate 3D Hand Pose from Single RGB Imagescited by 234
  5. 20193D Hand Shape and Pose from Images in the Wildcited by 190
  6. 20193D Hand Shape and Pose Estimation from a Single RGB Imagecited by 174
  7. 2018QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehensioncited by 164
  8. 2017GANerated Hands for Real-time 3D Hand Tracking from Monocular RGBcited by 147
  9. 2020Weakly-Supervised Mesh-Convolutional Hand Reconstruction in the Wildcited by 146
  10. 2015Capturing Hands in Action using Discriminative Salient Points and Physics Simulationcited by 115
  11. 2019Pushing the Envelope for RGB-based Dense 3D Hand Pose Estimation via Neural Renderingcited by 115
  12. 2018Hand Pose Estimation via Latent 2.5D Heatmap Regressioncited by 111
  13. 2020Monocular Real-time Hand Shape and Motion Capture using Multi-modal Datacited by 106
  14. 2017V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Mapcited by 102
  15. 2020Weakly Supervised 3D Hand Pose Estimation via Biomechanical Constraintscited by 97
  16. 2018Cross-modal Deep Variational Hand Pose Estimationcited by 89
  17. 2019A2J: Anchor-to-Joint Regression Network for 3D Articulated Pose Estimation from a Single Depth Imagecited by 56
  18. 2020DeepHandMesh: A Weakly-supervised Deep Encoder-Decoder Framework for High-fidelity Hand Mesh Modelingcited by 49
  19. 2021Towards Accurate Alignment in Real-time 3D Hand-Mesh Reconstructioncited by 44
  20. 2021Hand Image Understanding via Deep Multi-Task Learningcited by 31
  21. 2021Sign Language Production: A Reviewcited by 26
  22. 2020HandVoxNet: Deep Voxel-Based Network for 3D Hand Shape and Pose Estimation from a Single Depth Mapcited by 23
  23. 2019Generalized Feedback Loop for Joint Hand-Object Pose Estimationcited by 22
  24. 2017Depth-Based 3D Hand Pose Estimation: From Current Achievements to Future Goalscited by 21
  25. 2018Point-to-Pose Voting based Hand Pose Estimation using Residual Permutation Equivariant Layercited by 21
  26. 2015Hands Deep in Deep Learning for Hand Pose Estimationcited by 20
  27. 2020JGR-P2O: Joint Graph Reasoning based Pixel-to-Offset Prediction Network for 3D Hand Pose Estimation from a Single Depth Imagecited by 20
  28. 2017DeepPrior++: Improving Fast and Accurate 3D Hand Pose Estimationcited by 19
  29. 2017Dense 3D Regression for Hand Pose Estimationcited by 18
  30. 2018Analysis of Hand Segmentation in the Wildcited by 18
  31. 2020AWR: Adaptive Weighting Regression for 3D Hand Pose Estimationcited by 17
  32. 2021HandFoldingNet: A 3D Hand Pose Estimation Network Using Multiscale-Feature Guided Folding of a 2D Hand Skeletoncited by 16
  33. 2016Real-Time Hand Tracking Using a Sum of Anisotropic Gaussians Modelcited by 15
  34. 2018Augmented Skeleton Space Transfer for Depth-based Hand Pose Estimationcited by 15
  35. 2017Pose Guided Structured Region Ensemble Network for Cascaded Hand Pose Estimationcited by 12
  36. 2016Model-based Deep Hand Pose Estimationcited by 11
  37. 2017Region Ensemble Network: Improving Convolutional Network for Hand Pose Estimationcited by 11
  38. 2017Towards Good Practices for Deep 3D Hand Pose Estimationcited by 11

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 4; 2025: 1; 2026: 1.

7 papers.

  1. 2024Leveraging Pretrained Latent Representations for Few-Shot Imitation Learning on a Dexterous Robotic Handcited by 3
  2. 2023Tracking Fast by Learning Slow: An Event-based Speed Adaptive Hand Tracker Leveraging Knowledge in RGB Domaincited by 2
  3. 2025mmEgoHand: Egocentric Hand Pose Estimation and Gesture Recognition with Head-mounted Millimeter-wave Radar and IMUcited by 2
  4. 2024A Transformer Model for Boundary Detection in Continuous Sign Languagecited by 1
  5. 2024DOR3D-Net: Dense Ordinal Regression Network for 3D Hand Pose Estimationcited by 0
  6. 2024HandS3C: 3D Hand Mesh Reconstruction with State Space Spatial Channel Attention from RGB imagescited by 0
  7. 2026KAD-Net: Kinematics-Aware Decoupled Learning for Robust 3D Hand Pose Estimation from a Single Depth Imagecited by 0