mlsanityData to 14 Oct 2026

Kinematics-Induced Multimodal 3D Human Pose Estimation with Subject-Level Privacy

arXiv 2610.02943 ↗First posted 2 Oct 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Multimodal 3D Human Pose Estimation (3D HPE) combines complementary information from RGB, LiDAR, and mmWave radar, but models trained on correlated observations from the same individuals, raise privacy risks overlooked by record level analysis. We present a unified framework for multimodal 3D HPE that couples kinematics-induced sensor fusion with subject level privacy auditing and private training. First, our multimodal model aligns modality specific joint representation, injects skeletal structure and adaptively aggregates complementary sensor evidence for accurate pose prediction. Second, we formulate a black-box subject membership inference attack for 3D HPE, complemented by an empirical pointwise maximal leakage analysis, which characterizes how individual attack score outcomes change inference about the membership outcome. Third, we instantiate user-level differential privacy via Action Temporal Stratification, a population weighted within-subject sampling strategy that enforces action and temporal coverage. We evaluate our framework on the MM-Fi dataset across three diverse experimental protocols. Source-code will be released upon acceptance.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 41 of its 56 references that are in the map. First posted on arXiv 2 Oct 2026.

Builds on

41 of 56 references.

References · 41

  1. 2020Point Transformercited by 1,470
  2. 2017Renyi Differential Privacycited by 650
  3. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 403
  4. 2022Point Transformer V2: Grouped Vector Attention and Partition-based Poolingcited by 388
  5. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 272
  6. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 221
  7. 2019Differentially Private Learning with Adaptive Clippingcited by 220
  8. 2022MotionBERT: A Unified Perspective on Learning Human Motion Representationscited by 193
  9. 2022MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Videocited by 159
  10. 2019Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Imagecited by 132
  11. 20163D Human Pose Estimation = 2D Pose Estimation + Matchingcited by 113
  12. 2021TokenPose: Learning Keypoint Tokens for Human Pose Estimationcited by 107
29 more
  1. 2022DiffPose: Toward More Reliable 3D Pose Estimationcited by 100
  2. 2020Bypassing the Ambient Dimension: Private SGD with Gradient Subspace Identificationcited by 90
  3. 2021Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flowscited by 85
  4. 2023Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregationcited by 82
  5. 2023MM-Fi: Multi-Modal Non-Intrusive 4D Human Dataset for Versatile Wireless Sensingcited by 79
  6. 2018Attentive Relational Networks for Mapping Images to Scene Graphscited by 73
  7. 2019mm-Pose: Real-Time Human Skeletal Posture Estimation using mmWave Radars and CNNscited by 59
  8. 2022Mixed Differential Privacy in Computer Visioncited by 54
  9. 2022SoK: Let the Privacy Games Begin! A Unified Treatment of Data Inference Privacy in Machine Learningcited by 51
  10. 2021Multi-modal 3D Human Pose Estimation with 2D Weak Supervision in Autonomous Drivingcited by 40
  11. 2022mRI: Multi-modal 3D Human Pose Estimation Dataset using mmWave, RGB-D, and Inertial Sensorscited by 38
  12. 2024KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimationcited by 34
  13. 2022LiDARCap: Long-range Marker-less 3D Human Motion Capture with LiDAR Point Cloudscited by 33
  14. 2024Mind the Privacy Unit! User-Level Differential Privacy for Language Model Fine-Tuningcited by 24
  15. 2021Membership Inference Attacks on Deep Regression Models for Neuroimagingcited by 22
  16. 2024Diffusion Model is a Good Pose Estimator from 3D RF-Visioncited by 22
  17. 2014Accelerating Minibatch Stochastic Gradient Descent using Stratified Samplingcited by 21
  18. 2022Subject Membership Inference Attacks in Federated Learningcited by 20
  19. 2020HPERL: 3D Human Pose Estimation from RGB and LiDARcited by 19
  20. 2021Variance Reduced Training with Stratified Sampling for Forecasting Modelscited by 14
  21. 2024X-Fi: A Modality-Invariant Foundation Model for Multimodal Human Sensingcited by 13
  22. 2025Towards Balanced Multi-Modal Learning in 3D Human Pose Estimationcited by 12
  23. 2018Local Obfuscation Mechanisms for Hiding Probability Distributionscited by 9
  24. 2024ProbRadarM3F: mmWave Radar based Human Skeletal Pose Estimation with Probability Map Guided Multi-Format Feature Fusioncited by 5
  25. 2024Pre-training a Density-Aware Pose Transformer for Robust LiDAR-based 3D Human Pose Estimationcited by 5
  26. 2023Reconstructing 3D Human Pose from RGB-D Data with Occlusionscited by 3
  27. 2024On the Privacy of Selection Mechanisms with Gaussian Noisecited by 3
  28. 2019Understanding the efficacy, reliability and resiliency of computer vision techniques for malware detection and future research directionscited by 2
  29. 2025Differentially Private 2D Human Pose Estimationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimation
  2. 2023DiffHPE: Robust, Coherent 3D Human Pose Lifting with Diffusion
  3. 2026Privacy-Preserving Full-Body Meshing from mmWave Radar via Mesh Foundation Model Supervision
  4. 2025MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation
  5. 2026mmHRI: Towards Privacy-Preserving Human-Robot Interaction with Millimeter-Wave Radar
  6. 2024Pose Priors from Language Models
  7. 2026Flexible Geometric Guidance for Probabilistic Human Pose Estimation with Diffusion Models
  8. 2026TSR-Ego: Temporally Guided Stereo Refinement Framework for Egocentric 3D Human Pose Estimation