# LatentStealth: Unnoticeable and Efficient Adversarial Attacks on Expressive Human Pose and Shape Estimation

arXiv: https://arxiv.org/abs/2505.12009 · HTML: https://mlsanity.com/paper/2505.12009/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 44 references that are in the map.
- First posted on arXiv 17 May 2025.

## Builds on (most cited first, 30 of 35)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Expressive Body Capture: 3D Hands, Face, and Body from a Single Image](https://mlsanity.com/paper/1904.05866/) (2019)
- [HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation](https://mlsanity.com/paper/2011.14672/) (2020)
- [Simple Black-box Adversarial Attacks](https://mlsanity.com/paper/1905.07121/) (2019)
- [Learning to Estimate 3D Human Pose and Shape from a Single Color Image](https://mlsanity.com/paper/1805.04092/) (2018)
- [Whole-Body Human Pose Estimation in the Wild](https://mlsanity.com/paper/2007.11858/) (2020)
- [SMPLer-X: Scaling Up Expressive Human Pose and Shape Estimation](https://mlsanity.com/paper/2309.17448/) (2023)
- [PyMAF-X: Towards Well-aligned Full-body Model Regression from Monocular Images](https://mlsanity.com/paper/2207.06400/) (2022)
- [One-Stage 3D Whole-Body Mesh Recovery with Component Aware Transformer](https://mlsanity.com/paper/2303.16160/) (2023)
- [Adversarial Texture for Fooling Person Detectors in the Physical World](https://mlsanity.com/paper/2203.03373/) (2022)
- [Diffusion Models for Imperceptible and Transferable Adversarial Attack](https://mlsanity.com/paper/2305.08192/) (2023)
- [Universal Physical Camouflage Attacks on Object Detectors](https://mlsanity.com/paper/1909.04326/) (2019)
- [Accurate 3D Hand Pose Estimation for Whole-Body 3D Human Mesh Estimation](https://mlsanity.com/paper/2011.11534/) (2020)
- [Content-based Unrestricted Adversarial Attack](https://mlsanity.com/paper/2305.10665/) (2023)
- [Monocular Real-time Full Body Capture with Inter-part Correlations](https://mlsanity.com/paper/2012.06087/) (2020)
- [Global-to-Local Modeling for Video-based 3D Human Pose and Shape Estimation](https://mlsanity.com/paper/2303.14747/) (2023)
- [EventHPE: Event-based 3D Human Pose and Shape Estimation](https://mlsanity.com/paper/2108.06819/) (2021)
- [Can Push-forward Generative Models Fit Multimodal Distributions?](https://mlsanity.com/paper/2206.14476/) (2022)
- [Benchmarking and Analyzing 3D Human Pose and Shape Estimation Beyond Algorithms](https://mlsanity.com/paper/2209.10529/) (2022)
- [Towards Transferable Targeted 3D Adversarial Attack in the Physical World](https://mlsanity.com/paper/2312.09558/) (2023)
- [Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attack](https://mlsanity.com/paper/2103.05347/) (2021)
- [Garment4D: Garment Reconstruction from Point Cloud Sequences](https://mlsanity.com/paper/2112.04159/) (2021)
- [A Comprehensive Study of the Robustness for LiDAR-based 3D Object Detectors against Adversarial Attacks](https://mlsanity.com/paper/2212.10230/) (2022)
- [Distracting Downpour: Adversarial Weather Attacks for Motion Estimation](https://mlsanity.com/paper/2305.06716/) (2023)
- [LatentPoison - Adversarial Attacks On The Latent Space](https://mlsanity.com/paper/1711.02879/) (2017)
- [Generating Adversarial Attacks in the Latent Space](https://mlsanity.com/paper/2304.04386/) (2023)

## Abstract (arXiv)

Expressive human pose and shape estimation (EHPS) plays a central role in digital human generation, particularly in live-streaming applications. However, most existing EHPS models focus primarily on minimizing estimation errors, with limited attention on potential security vulnerabilities, such as generating inappropriate content, violent actions, or racially offensive gestures and expressions. Current adversarial attacks on EHPS models often generate visually conspicuous perturbations, limiting their practicality and ability to expose real-world security threats. To address this limitation, we propose an unnoticeable adversarial method, termed LatentStealth, specifically tailored for EHPS models. The key idea is to exploit the structured latent representations of natural images as the medium for crafting perturbations. Instead of injecting noise directly into the pixel space, our method projects inputs into the latent space, where adversarial patterns are generated and progressively refined along optimized directions. This latent-space manipulation enables the attack to maintain high imperceptibility while preserving its effectiveness. Furthermore, as the optimization process is guided by only a small number of model output queries, the framework achieves competitive attack performance with low computational overhead, making it both practical and efficient for real-world scenarios. Extensive experiments on the 3DPW and UBody datasets demonstrate the superiority of LatentStealth, revealing critical vulnerabilities in current systems. These findings highlight the urgent need to address and mitigate security risks in digital human generation technologies.