# Gesture-Aware Pretraining and Token Fusion for 3D Hand Pose Estimation

arXiv: https://arxiv.org/abs/2603.17396 · HTML: https://mlsanity.com/paper/2603.17396/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 25 references that are in the map.
- First posted on arXiv 18 Mar 2026.

## Builds on (most cited first, 20 of 20)

- [Deep High-Resolution Representation Learning for Visual Recognition](https://mlsanity.com/paper/1908.07919/) (2019)
- [Deep High-Resolution Representation Learning for Human Pose Estimation](https://mlsanity.com/paper/1902.09212/) (2019)
- [On the Continuity of Rotation Representations in Neural Networks](https://mlsanity.com/paper/1812.07035/) (2018)
- [End-to-end Recovery of Human Shape and Pose](https://mlsanity.com/paper/1712.06584/) (2017)
- [Reconstructing Hands in 3D with Transformers](https://mlsanity.com/paper/2312.05251/) (2023)
- [FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Images](https://mlsanity.com/paper/1909.04349/) (2019)
- [InterHand2.6M: A Dataset and Baseline for 3D Interacting Hand Pose Estimation from a Single RGB Image](https://mlsanity.com/paper/2008.09309/) (2020)
- [High-Resolution Representations for Labeling Pixels and Regions](https://mlsanity.com/paper/1904.04514/) (2019)
- [3D Hand Shape and Pose from Images in the Wild](https://mlsanity.com/paper/1902.03451/) (2019)
- [3D Hand Shape and Pose Estimation from a Single RGB Image](https://mlsanity.com/paper/1903.00812/) (2019)
- [FrankMocap: A Monocular 3D Whole-Body Pose Estimation System via Regression and Integration](https://mlsanity.com/paper/2108.06428/) (2021)
- [WiLoR: End-to-end 3D Hand Localization and Reconstruction in-the-wild](https://mlsanity.com/paper/2409.12259/) (2024)
- [Monocular Real-time Hand Shape and Motion Capture using Multi-modal Data](https://mlsanity.com/paper/2003.09572/) (2020)
- [Interacting Attention Graph for Single Image Two-Hand Reconstruction](https://mlsanity.com/paper/2203.09364/) (2022)
- [Model-based 3D Hand Reconstruction via Self-Supervised Learning](https://mlsanity.com/paper/2103.11703/) (2021)
- [ACR: Attention Collaboration-based Regressor for Arbitrary Two-Hand Reconstruction](https://mlsanity.com/paper/2303.05938/) (2023)
- [Im2Hands: Learning Attentive Implicit Representation of Interacting Two-Hand Shapes](https://mlsanity.com/paper/2302.14348/) (2023)
- [BiHand: Recovering Hand Mesh with Multi-stage Bisected Hourglass Networks](https://mlsanity.com/paper/2008.05079/) (2020)
- [SHuBERT: Self-Supervised Sign Language Representation Learning via Multi-Stream Cluster Prediction](https://mlsanity.com/paper/2411.16765/) (2024)
- [Extract-and-Adaptation Network for 3D Interacting Hand Mesh Recovery](https://mlsanity.com/paper/2309.01943/) (2023)

## Abstract (arXiv)

Estimating 3D hand pose from monocular RGB images is fundamental for applications in AR/VR, human-computer interaction, and sign language understanding. In this work we focus on a scenario where a discrete set of gesture labels is available and show that gesture semantics can serve as a powerful inductive bias for 3D pose estimation. We present a two-stage framework: gesture-aware pretraining that learns an informative embedding space using coarse and fine gesture labels from InterHand2.6M, followed by a per-joint token Transformer guided by gesture embeddings as intermediate representations for final regression of MANO hand parameters. Training is driven by a layered objective over parameters, joints, and structural constraints. Experiments on InterHand2.6M demonstrate that gesture-aware pretraining consistently improves single-hand accuracy over the state-of-the-art EANet baseline, and that the benefit transfers across architectures without any modification.