# ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization

arXiv: https://arxiv.org/abs/2505.10250 · HTML: https://mlsanity.com/paper/2505.10250/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 53 of its 65 references that are in the map.
- First posted on arXiv 15 May 2025.

## Builds on (most cited first, 30 of 53)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Expressive Body Capture: 3D Hands, Face, and Body from a Single Image](https://mlsanity.com/paper/1904.05866/) (2019)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [End-to-end Recovery of Human Shape and Pose](https://mlsanity.com/paper/1712.06584/) (2017)
- [Diffusion Model Alignment Using Direct Preference Optimization](https://mlsanity.com/paper/2311.12908/) (2023)
- [Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop](https://mlsanity.com/paper/1909.12828/) (2019)
- [Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision](https://mlsanity.com/paper/1611.09813/) (2016)
- [Aligning Text-to-Image Models using Human Feedback](https://mlsanity.com/paper/2302.12192/) (2023)
- [Humans in 4D: Reconstructing and Tracking Humans with Transformers](https://mlsanity.com/paper/2305.20091/) (2023)
- [Reward learning from human preferences and demonstrations in Atari](https://mlsanity.com/paper/1811.06521/) (2018)
- [Learning 3D Human Dynamics from Video](https://mlsanity.com/paper/1812.01601/) (2018)
- [HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation](https://mlsanity.com/paper/2011.14672/) (2020)
- [PyMAF: 3D Human Pose and Shape Regression with Pyramidal Mesh Alignment Feedback Loop](https://mlsanity.com/paper/2103.16507/) (2021)
- [BEDLAM: A Synthetic Dataset of Bodies Exhibiting Detailed Lifelike Animated Motion](https://mlsanity.com/paper/2306.16940/) (2023)
- [Unite the People: Closing the Loop Between 3D and 2D Human Representations](https://mlsanity.com/paper/1701.02468/) (2017)
- [HIVE: Harnessing Human Feedback for Instructional Visual Editing](https://mlsanity.com/paper/2303.09618/) (2023)
- [SMPLer-X: Scaling Up Expressive Human Pose and Shape Estimation](https://mlsanity.com/paper/2309.17448/) (2023)
- [Probabilistic Modeling for Human Mesh Recovery](https://mlsanity.com/paper/2108.11944/) (2021)
- [SPEC: Seeing People in the Wild with an Estimated Camera](https://mlsanity.com/paper/2110.00620/) (2021)
- [One-Stage 3D Whole-Body Mesh Recovery with Component Aware Transformer](https://mlsanity.com/paper/2303.16160/) (2023)
- [DNA-Rendering: A Diverse Neural Actor Repository for High-Fidelity Human-centric Rendering](https://mlsanity.com/paper/2307.10173/) (2023)
- [TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation](https://mlsanity.com/paper/2404.16752/) (2024)
- [Hi4D: 4D Instance Segmentation of Close Human Interaction](https://mlsanity.com/paper/2303.15380/) (2023)

## Built on by (most cited first, 7 of 7)

- [Puppeteer: Rig and Animate Your 3D Models](https://mlsanity.com/paper/2508.10898/) (2025)
- [From Editor to Dense Geometry Estimator](https://mlsanity.com/paper/2509.04338/) (2025)
- [VLM-Guided Group Preference Alignment for Diffusion-based Human Mesh Recovery](https://mlsanity.com/paper/2602.19180/) (2026)
- [MotionGRPO: Overcoming Low Intra-Group Diversity in GRPO-Based Egocentric Motion Recovery](https://mlsanity.com/paper/2605.05680/) (2026)
- [On the Use of Hierarchical Vision Foundation Models for Low-Cost Human Mesh Recovery and Pose Estimation](https://mlsanity.com/paper/2510.12660/) (2025)
- [AnimaSpark: A Feed-Forward Method for Animating Arbitrary 3D Objects](https://mlsanity.com/paper/2606.10988/) (2026)
- [DreamHMR: From Video Generation to Human Motion Modeling via Structural Focusing](https://mlsanity.com/paper/2610.20124/) (2026)

## Abstract (arXiv)

Human mesh recovery (HMR) from a single image is inherently ill-posed due to depth ambiguity and occlusions. Probabilistic methods have tried to solve this by generating numerous plausible 3D human mesh predictions, but they often exhibit misalignment with 2D image observations and weak robustness to in-the-wild images. To address these issues, we propose ADHMR, a framework that Aligns a Diffusion-based HMR model in a preference optimization manner. First, we train a human mesh prediction assessment model, HMR-Scorer, capable of evaluating predictions even for in-the-wild images without 3D annotations. We then use HMR-Scorer to create a preference dataset, where each input image has a pair of winner and loser mesh predictions. This dataset is used to finetune the base model using direct preference optimization. Moreover, HMR-Scorer also helps improve existing HMR models by data cleaning, even with fewer training samples. Extensive experiments show that ADHMR outperforms current state-of-the-art methods. Code is available at: https://github.com/shenwenhao01/ADHMR.