Pith. sign in

REVIEW 5 major objections 6 minor 76 references

ChairPose: Pressure-based Chair Morphology Grounded Sitting Pose Estimation through Simulation-Assisted Training

T0 review · 5 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that a thin pressure mattress plus a one-time 3D scan of the chair can reconstruct full-body seated pose on users and chairs never seen during training, with a mean per-joint position error of 89.4 mm.

desk verdict Novel and well-evaluated pressure-to-pose system whose absolute accuracy is undercut by monocular pseudo-ground-truth; deserves refereeing but needs independent validation. read the letter →

arxiv 2508.01850 v1 pith:XGLLVF5O submitted 2025-08-03 cs.HC cs.AI

Lala Shakti Swarup Ray , Vitor Fortes Rey , Bo Zhou , Paul Lukowicz , Sungho Suh This is my paper · ORCID
classification cs.HCcs.AI
keywords seatedposeestimationpressuresensingchairmorphologymotionquantizationVQ-VAEphysics-basedsimulationdataaugmentationprivacy-preserving
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

ChairPose sets out to prove that full-body 3D seated pose can be reconstructed passively: a thin pressure mattress placed on a chair, together with a one-time scan of that chair, replaces cameras and wearable sensors. The paper claims this works on chairs and people the model has never seen, reporting a mean per-joint position error of 89.4 mm when both are held out. The system's two-stage design first quantizes body motion into discrete tokens, then predicts those tokens autoregressively from pressure dynamics and chair geometry, and a physics-based simulation pipeline supplies extra training pairs across 12 synthetic chair shapes. If the claim holds, continuous posture monitoring, center-of-mass estimation, and seated activity recognition become possible in privacy-sensitive settings with no cameras or body-worn sensors.

What carries the argument

The load-bearing mechanism is the two-stage generative pipeline. MotionQuantizer is a VQ-VAE with a U-Net backbone that compresses n-second chunks of SMPL joint positions and motion derivatives into codebook tokens, using exponential moving average updates and quantization dropout to keep the codebook stable. Pressure2Pose is an autoregressive classifier over those tokens, conditioned on an 80x28 pressure-map sequence, a 5000-point chair point cloud encoded by PointNet, and the previously predicted token, with a learned start token to begin generation; outputs are decoded back into continuous SMPL joint positions. The second mechanism is the simulation-assisted augmentation pipeline, which replaces a flat-floor pressure simulator with 12 chair CAD models and runs ragdoll physics on motion-capture sequences so the synthetic pressure-pose pairs obey plausible body-chair contact.

What would settle it

Run a marker-based or multi-view motion-capture system on the same 12 activities across the four chairs, and compare ChairPose's predicted joints to those marker positions for held-out users and chairs; if the mean per-joint error against true marker-based ground truth is substantially above 89.4 mm, especially for arm and head joints, the central claim fails. A second direct test is to place the mattress on a soft, deformable seat and see whether the claimed chair-agnostic accuracy collapses.

Watch

Extended reading notes

Core claim

The core claim is that pressure maps alone are a sufficient signal for continuous full-body 3D seated pose estimation, provided the model is told the chair's 3D shape and trained with physically grounded synthetic examples. The paper demonstrates this with a two-stage generative architecture whose MotionQuantizer turns SMPL pose sequences into a discrete codebook, and whose Pressure2Pose module classifies each next codebook token from pressure frames, a PointNet-encoded chair point cloud, and the previously predicted token. The reported results are 53.5 mm MPJPE for unseen users, 88.2 mm for unseen chairs, and 89.4 mm for unseen users and chairs together, down from 260.7 mm for a direct-regression baseline under the hardest protocol. The same reconstructed mesh yields clinically oriented outputs: volumetric center-of-mass estimation with lower error than direct regression, spine-angle monitoring with mean errors under 4.2 degrees, and activity recognition improved by fusing the predicted pose with raw pressure.

Load-bearing premise

The reported accuracy numbers assume the monocular vision reconstructions used as ground truth are faithful for seated and partially occluded bodies; any systematic bias in those reconstructions would directly shift every reported error.

Editorial extensions

If this is right

  • If the reported numbers hold, a single flexible mattress plus a 3D chair scan can replace cameras and IMUs for seated pose monitoring, working on chairs and users the model was never trained on.
  • The full-body SMPL mesh it outputs supports downstream tasks directly: center-of-mass estimation, spine angle monitoring, and seated action recognition all improve over pressure-only or direct-regression baselines.
  • The fixed roughly two-second latency of the accurate mode suits offline clinical or ergonomic assessment, while the fast baseline mode serves interactive, latency-critical uses at lower accuracy.
  • Physics-based synthetic augmentation, not simple image-style augmentation, is what drives generalization: the paper reports error reductions up to 46 percent from adding simulated pressure-pose pairs to the training set.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Extrapolating from the paper's four chair types, the strongest untested boundary is deformable seating: soft cushions and dynamic surfaces change pressure patterns, so putting the mattress on a plush sofa would be the natural next experiment.
  • Because the ground truth comes from monocular vision extraction rather than marker-based capture, the 89.4 mm figure measures agreement with that vision model; a multi-view or marker-based validation would separate sensor ambiguity from pseudo-ground-truth bias.
  • The synthetic data pipeline is a reusable asset: the same ragdoll-plus-pressure simulation could generate training pairs for other chair-embedded tasks such as fall detection, occupancy sensing, or driver monitoring without new real data collection.
  • The tokenization approach suggests a path to cross-modal supervision: if pressure-to-token classification works, the same codebook could accept pose context from text or audio, enabling pose generation from multiple sparse modalities.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper introduces ChairPose, a pressure-mat-based system for full-body 3D seated pose estimation. It uses a two-stage generative pipeline: a VQ-VAE motion quantizer that discretizes SMPL pose sequences, and an autoregressive token predictor conditioned on pressure maps, a 3D chair point cloud, and previous pose tokens. Training is augmented by a physics-based simulation pipeline (PresSim plus ragdoll dynamics) that generates synthetic pressure-pose pairs for 12 additional chair models. The system is evaluated on a new dataset (TDSD) of 8 participants, 4 chairs, and 12 seated activities, under leave-one-user-out (LOUOCV), leave-one-chair-out (LOCOCV), and leave-one-(user+chair)-out (LOCUOCV) protocols. Reported MPJPE is 53.5 mm (LOUOCV), 88.2 mm (LOCOCV), and 89.4 mm (LOCUOCV), with large improvements over the non-quantized baseline and over re-implemented pressure-based baselines. Additional experiments cover ablations of the motion quantizer, data augmentation, volumetric center-of-mass estimation, posture monitoring, and activity recognition.

Significance. If the accuracy numbers are taken at face value, ChairPose is a substantial step for privacy-preserving seated pose estimation: it removes wearables and cameras, incorporates chair geometry explicitly, and demonstrates generalization to unseen users and chairs through physics-based simulated augmentation. The paper ships publicly available code and data, a sensible cross-validation protocol, and a detailed ablation study. The reported gains from motion quantization and from simulation-assisted augmentation are large and internally consistent across the three cross-validation settings. However, the central generalization claim rests on the validity of the pose ground truth, which is currently a monocular vision model output, and the paper overstates the sensor-side independence by requiring a 3D chair scan as input.

major comments (5)
  1. [§4.3, Tables 5–7] The ground-truth poses are produced by SMPLer-X from a single iPhone TrueDepth camera fixed at a 45-degree front-left viewpoint, and no independent validation is reported for seated configurations. Chair-induced occlusions of the lower legs, feet, and parts of the thighs are precisely where monocular SMPLer-X is most likely to have systematic bias, and any such bias directly shifts every reported MPJPE/PA-MPJPE/MPVE value. The cited DMCB benchmark (refs. [43,44]) concerned loose garments and does not establish SMPLer-X accuracy for seated poses with chair occlusion. Please add an independent validation subset, such as optical marker-based motion capture, multi-view triangulation, or a manually annotated joint sample, and report per-joint errors so the pseudo-ground-truth bias can be assessed.
  2. [§5.1.2, Tables 5–7] SMPLer-X is used both as the source of ground-truth annotations (Section 4.3) and as a comparison method in the evaluation tables, where it reports about 11 mm error. This makes the ~11 mm row a self-consistency measure rather than an independent accuracy benchmark, and it conflates the reference with a compared system. Either remove SMPLer-X from the comparison tables or explicitly relabel its numbers as pseudo-ground-truth agreement, not as an external reference.
  3. [Table 7] In LOCUOCV, PA-MPJPE (88.7 mm) is nearly identical to MPJPE (89.4 mm), whereas in LOUOCV PA-MPJPE is about 16 mm lower than MPJPE (37.9 vs. 53.5 mm) and in LOCOCV it is about 11 mm lower (77.1 vs. 88.2 mm). Such near-equality after Procrustes alignment is unusual and suggests a non-rigid per-joint bias that is not captured by a global similarity transform, or an artifact of the shared pseudo-ground-truth. Please report per-joint errors, the fitted Procrustes parameters, or a scatter analysis for this condition to explain why alignment does not reduce the error.
  4. [Abstract, §3.2, §4.1] The paper claims that the system operates 'solely on pressure sensing' and 'independently of chair geometry,' yet the P2P module takes as input the chair point cloud M ∈ R^{5000×3} (Section 3.2) and Section 4.1 describes acquiring 3D chair scans with a Ferret Pro scanner. This is an overclaim: the system is chair-geometry-aware and requires a scan at inference time, even though it does not require per-chair retraining. Please revise the abstract, the contribution list, and Table 1 to distinguish 'no per-chair retraining' from 'no chair-geometry input.'
  5. [§5.1.2 (data augmentation ablation)] The simulation-assisted augmentation is a central contribution, but the paper does not report any direct measure of sim-to-real transfer, such as pressure-map distribution similarity or a real-synthetic domain gap analysis. The reported accuracy gains are consistent across cross-validation protocols, which is encouraging, but an explicit validation that synthetic pressure maps resemble real sensor readings on the four physical chairs would strengthen the claim that the physics-based pipeline is responsible for the improvement rather than merely adding data diversity.
minor comments (6)
  1. [§5.1.1, Eq. (10)] The metric notation in Eq. (10) is internally inconsistent: 'X is \hat{p}_j' is not a proper definition, and the norm subscript is missing. Please rewrite the equation with consistent notation for prediction and ground truth.
  2. [§4.5, Table 3] The codebook size is given as 1028, which is not a power of two and is an unusual choice; please confirm whether this is a typo for 1024 and, if not, explain the rationale.
  3. [§7.3, Figure 9] The text states that the MQ+sequence mode has a fixed ~2 second latency 'regardless of input length,' but the source of this fixed delay is not explained given the 1-second quantization window; please clarify whether the delay comes from the autoregressive decoding, the sequence loss, or the temporal receptive field.
  4. [§7.2.3] Error accumulation is acknowledged as a limitation of the autoregressive decoder, but no quantitative drift analysis is provided; an error-over-time plot would make the severity of this issue concrete.
  5. [Table 1] The column 'Cross-surface generalization' is not defined in the table or the text; please clarify whether it indicates operation across chair shapes, planar surfaces only, or something else.
  6. [§6.2] The posture-monitoring experiment reports mean absolute errors of 4.2° for lumbar flexion and 3.8° for thoracic tilt, but the paper does not specify how the ground-truth angles were computed from the SMPL poses or how the predicted angles are derived; please add this detail for reproducibility.

Circularity Check

0 steps flagged · score 2.0 of 10

No construction-level circularity: the pressure-to-pose model is trained and evaluated on held-out real data; the pseudo-ground-truth and self-cited simulator concerns are validity risks, not derivation equivalence.

full rationale

The central chain is pressure map plus chair scan to tokenized SMPL pose. The P2P model is trained on a mixture of real TDSD samples and PresSim-generated synthetic samples, and all generalization numbers (LOUOCV, LOCOCV, LOCUOCV) are computed on real held-out subjects and chairs. No fitted parameter is reused as a prediction target, and no equation defining the output is identical to an input by construction. The main caveats -- SMPLer-X pseudo-ground-truth from a single 45-degree iPhone camera (Sections 4.2-4.3) and the same SMPLer-X model appearing as a comparison baseline (Tables 5-7) -- undermine the interpretation of the absolute error as true physical pose error, but they do not make the pressure-to-pose mapping equivalent to its training labels; the mapping is still tested on unseen real data. Self-citations to PresSim [42] and DMCB [43,44] are load-bearing for the simulator and for the pseudo-GT justification, but both are externally checkable artifacts (published frameworks and benchmarks) rather than unverified assertions, and the simulator's utility is independently tested by the real-data evaluation. Therefore circularity is low; the dominant concern is ground-truth validity, not circular derivation.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central results rest mainly on empirical domain assumptions about sensor informativeness, simulator fidelity, and pseudo-ground-truth quality, plus a set of hand-chosen hyperparameters. There are no invented physical entities such as new particles or forces. The free parameters are standard VQ-VAE and training hyperparameters, but codebook size and window length are selected via ablation and influence the central accuracy numbers.

free parameters (5)
  • codebook size C = 1028
    Selected by ablation on FID and R-Precision (Table 3); all downstream MPJPE numbers depend on this discrete motion vocabulary.
  • quantization window n = 1 second
    Selected by the same ablation; defines the temporal receptive field and the fixed roughly 2-second latency of the high-accuracy mode.
  • sequence loss weight lambda = 0.5
    Hand-set balance between token cross-entropy and decoded-pose MSE in Equation 7; directly affects temporal coherence and MPJPE.
  • quantization dropout probability p = 0.2
    Hand-set regularization in Equation 4; included in the best MQ configuration reported in Table 4.
  • EMA smoothing factor alpha = 0.99
    Hand-set codebook smoothing in Equation 3; part of the best MQ configuration and not swept separately.
assumptions (5)
  • domain assumption SMPLer-X pseudo-ground truth from a single iPhone TrueDepth camera is sufficiently accurate to serve as ground truth.
    Sections 4.2 and 4.3; all MPJPE numbers inherit this assumption, and no marker-based or multi-view validation is reported.
  • domain assumption PresSim's physics-based simulation generates pressure maps that transfer to real TPE pressure mats on non-planar chairs.
    Section 4.4; the synthetic pressure data are used for training, and transfer is only tested indirectly through downstream pose accuracy.
  • domain assumption Ragdoll-simulated seated motions are representative of real seated behavior.
    Section 4.4; these motions generate 1,153,500 synthetic frames used to train Pressure2Pose.
  • domain assumption An 80x28 pressure grid with the described contact area is sufficient to disambiguate full-body seated poses.
    The whole method depends on pressure informativeness; the paper's own limitation 7.2.2 acknowledges that arms and head are weakly observable.
  • domain assumption Tap-based session-boundary synchronization aligns pressure and pose streams accurately throughout each session.
    Section 4.2; taps only at the beginning and end of each session, with no reported per-frame drift analysis or continuous synchronization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ChairPose: Pressure-based Chair Morphology Grounded Sitting Pose Estimation through Simulation-Assisted Training." pith.science (2026). https://pith.science/paper/XGLLVF5O

@misc{pith2026250801850,
  author       = {Pith},
  title        = {Pith review of: ChairPose: Pressure-based Chair Morphology Grounded Sitting Pose Estimation through Simulation-Assisted Training},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XGLLVF5O}},
  note         = {Machine review of arXiv:2508.01850}
}
read the original abstract

Prolonged seated activity is increasingly common in modern environments, raising concerns around musculoskeletal health, ergonomics, and the design of responsive interactive systems. Existing posture sensing methods such as vision-based or wearable approaches face limitations including occlusion, privacy concerns, user discomfort, and restricted deployment flexibility. We introduce ChairPose, the first full body, wearable free seated pose estimation system that relies solely on pressure sensing and operates independently of chair geometry. ChairPose employs a two stage generative model trained on pressure maps captured from a thin, chair agnostic sensing mattress. Unlike prior approaches, our method explicitly incorporates chair morphology into the inference process, enabling accurate, occlusion free, and privacy preserving pose estimation. To support generalization across diverse users and chairs, we introduce a physics driven data augmentation pipeline that simulates realistic variations in posture and seating conditions. Evaluated across eight users and four distinct chairs, ChairPose achieves a mean per joint position error of 89.4 mm when both the user and the chair are unseen, demonstrating robust generalization to novel real world generalizability. ChairPose expands the design space for posture aware interactive systems, with potential applications in ergonomics, healthcare, and adaptive user interfaces.

Figures

Figures reproduced from arXiv: 2508.01850 by the authors.

Figure 1
Figure 1. ChairPose predicts full-body volumetric pose sequence from pressure inputs while accounting for chair morphology [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. ChairPose employs a two-step training architecture: first, the MQ module quantizes SMPL pose sequences into discrete [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Hardware setup: pressure mat placed on the chair [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: The 12 predefined seated actions used in data col [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Intelligent data augmentation pipeline that extends [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Visualization of the input pressure map, 3D chair point cloud, and predicted SMPL body pose, error color-map as [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: ChairPose’s full body seated pose estimation and [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: Example applications of ChairPose in real-world [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: Latency (startup delay) vs. MPJPE for different input [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

76 extracted references · 72 canonical work pages

  1. [1]

    Reem Alattas and Khaled Elleithy. 2014. Detecting and minimizing bad posture using postuino among engineering students. In 2014 2nd International Conference on Artificial Intelligence, Modelling and Simulation . IEEE, 344–349

    work page 2014

  2. [2]

    Taraneh Aminosharieh Najafi, Antonio Abramo, Kyandoghere Kyamakya, and Antonio Affanni. 2022. Development of a smart chair sensors system and classi- fication of sitting postures with deep learning algorithms. Sensors 22, 15 (2022), 5585

    work page 2022

  3. [3]

    Marilda Ardito, Fabiana Mascolo, Martina Valentini, and Francesco Dell’Olio

    work page

  4. [4]

    CMA Ashruf. 2002. Thin flexible pressure sensors. Sensor Review 22, 4 (2002), 322–327

    work page 2002

  5. [5]

    Daniele Bibbo, Marco Carli, Silvia Conforto, and Federica Battisti. 2019. A sitting posture monitoring instrument to assess different levels of cognitive engagement. Sensors 19, 3 (2019), 455

    work page 2019

  6. [6]

    Zhongang Cai, Wanqi Yin, Ailing Zeng, Chen Wei, Qingping Sun, Wang Yanjun, Hui En Pang, Haiyi Mei, Mingyuan Zhang, Lei Zhang, et al . 2023. Smpler-x: Scaling up expressive human pose and shape estimation. Advances in Neural Information Processing Systems 36 (2023), 11454–11468. ChairPose UIST ’25, September 28-October 1, 2025, Busan, Republic of Korea

    work page 2023

  7. [7]

    Sevim Celik, Kadir Celik, ELİF DİRİMEŞE, Nurten Taşdemir, Tarik Arik, and Ibrahim Buyukkara. 2018. Determination of pain in musculoskeletal system reported by office workers and the pain risk factors. International journal of occupational medicine and environmental health 31, 1 (2018)

    work page 2018

  8. [8]

    Kaixin Chen, Lin Zhang, Zhong Wang, Shengjie Zhao, and Yicong Zhou. 2025. Skeleton-aware graph-based adversarial networks for human pose estimation from sparse IMUs. ACM Transactions on Multimedia Computing, Communications and Applications 21, 4 (2025), 1–22

    work page 2025

Show all 76 references
  1. [9]

    Wenqiang Chen, Yexin Hu, Wei Song, Yingcheng Liu, Antonio Torralba, and Wo- jciech Matusik. 2024. CAvatar: Real-time Human Activity Mesh Reconstruction via Tactile Carpets. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 7, 4 (2024), 1–24

    2024

  2. [10]

    Wenshuo Chen, Hongru Xiao, Erhang Zhang, Lijie Hu, Lei Wang, Mengyuan Liu, and Chen Chen. 2024. SATO: Stable Text-to-Motion Framework. In Proceedings of the 32nd ACM International Conference on Multimedia . 6989–6997

    2024

  3. [11]

    Henry M Clever, Patrick L Grady, Greg Turk, and Charles C Kemp. 2022. Bodypressure-inferring body pose and contact pressure from a depth image. IEEE Transactions on Pattern Analysis and Machine Intelligence 45, 1 (2022), 137– 153

    2022

  4. [12]

    Hadi Daneshmandi, Alireza Choobineh, Haleh Ghaem, and Mehran Karimi. 2017. Adverse effects of prolonged sitting behavior on the general health of office workers. Journal of lifestyle medicine 7, 2 (2017), 69

    2017

  5. [13]

    Carrie Demmans, Sriram Subramanian, and Jon Titus. 2007. Posture monitoring and improvement for laptop use. In CHI’07 Extended Abstracts on Human Factors in Computing Systems. 2357–2362

    2007

  6. [14]

    Ke Fan, Junshu Tang, Weijian Cao, Ran Yi, Moran Li, Jingyu Gong, Jiangning Zhang, Yabiao Wang, Chengjie Wang, and Lizhuang Ma. 2024. Freemotion: A unified framework for number-free text-to-motion synthesis. In European Conference on Computer Vision . Springer, 93–109

    2024

  7. [15]

    Lin Feng, Ziyi Li, and Chen Liu. 2019. Are you sitting right?-sitting posture recog- nition using RF signals. In 2019 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM). IEEE, 1–6

    2019

  8. [16]

    Lin Feng, Ziyi Li, Chen Liu, Xiaojiang Chen, Xiao Yin, and Dingyi Fang. 2020. SitR: Sitting posture recognition using RF signals. IEEE Internet of Things Journal 7, 12 (2020), 11492–11504

    2020

  9. [17]

    Nabarun Goswami, Yusuke Mukuta, and Tatsuya Harada. 2024. Hypervq: Mlr- based vector quantization in hyperbolic space. arXiv preprint arXiv:2403.13015 (2024)

    2024 arXiv

  10. [18]

    Mikael Anne Greenwood-Hickman, Supun Nakandala, Marta M Jankowska, Dori E Rosenberg, Fatima Tuz-Zahra, John Bellettiere, Jordan Carlson, Paul R Hibbing, Jingjing Zou, Andrea Z Lacroix, et al. 2021. The CNN Hip Accelerometer Posture (CHAP) method for classifying sitting pattern...

    2021

  11. [19]

    Chuan Guo, Shihao Zou, Xinxin Zuo, Sen Wang, Wei Ji, Xingyu Li, and Li Cheng

  12. [20]

    Jongseong Gwak, Kazuyoshi Arata, Takumi Yamakawa, Hideo Tobata, Motoki Shino, and Yoshihiro Suda. 2024. Physiological Responses Related to Sitting Comfort Due to Changes in Seat Parameters. Applied Sciences 14, 17 (2024), 7870

    2024

  13. [21]

    Isaac Han, Seoyoung Lee, Sangyeon Park, Ecehan Akan, Yiyue Luo, and Kyung- Joong Kim. [n. d.]. Smart Insole: Predicting 3D human pose from foot pressure. In 2nd NeurIPS Workshop on Touch Processing: From Data to Knowledge

  14. [22]

    Jiawei Huang and Ding Zhou. 2024. A scalable real-time computer vision system for student posture detection in smart classrooms. Education and Information Technologies 29, 1 (2024), 917–937

    2024

  15. [23]

    Hang Jin, Xin He, Lingyun Wang, Yujun Zhu, Weiwei Jiang, and Xiaobo Zhou

  16. [24]

    Christian Krauter, Katrin Angerbauer, Aimée Sousa Calepso, Alexander Achberger, Sven Mayer, and Michael Sedlmair. 2024. Sitting posture recognition and feedback: a literature review. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems . 1–20

    2024

  17. [25]

    Linhan Li, Guanci Yang, Yang Li, Dongying Zhu, and Ling He. 2023. Abnormal sit- ting posture recognition based on multi-scale spatiotemporal features of skeleton graph. Engineering Applications of Artificial Intelligence 123 (2023), 106374

    2023

  18. [26]

    Weijian Li, Shijie Li, and Huaiguang Jiang. 2024. Semantically Encoding Enhance- ments for Unsupervised Color Feature Separation. In 2024 IEEE International Conference on Smart City (SmartCity) . IEEE, 77–82

    2024

  19. [27]

    Yanwei Li, Chengyao Wang, and Jiaya Jia. 2024. Llama-vid: An image is worth 2 tokens in large language models. In European Conference on Computer Vision . Springer, 323–340

    2024

  20. [28]

    Matthew Loper, Naureen Mahmood, Javier Romero, Gerard Pons-Moll, and Michael J Black. 2023. SMPL: A skinned multi-person linear model. In Seminal Graphics Papers: Pushing the Boundaries, Volume 2 . 851–866

    2023

  21. [29]

    Yiyue Luo, Yunzhu Li, Michael Foshey, Wan Shou, Pratyusha Sharma, Tomás Palacios, Antonio Torralba, and Wojciech Matusik. 2021. Intelligent Carpet: Inferring 3D Human Pose from Tactile Signals. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . 112...

    2021

  22. [30]

    Matthew Mclaughlin, AJ Atkin, L Starr, A Hall, L Wolfenden, R Sutherland, J Wiggers, A Ramirez, P Hallal, M Pratt, et al . 2020. Worldwide surveillance of self-reported sitting time: a scoping review. International Journal of Behavioral Nutrition and Physical Activity 17, 1 (2...

    2020

  23. [31]

    Teruhiro Mizumoto, Yasuhiro Otoda, Chihiro Nakajima, Mitsuhiro Kohana, Mo- tohiro Uenishi, Keiichi Yasumoto, and Yutaka Arakawa. 2020. Design and imple- mentation of sensor-embedded chair for continuous sitting posture recognition. IEICE TRANSACTIONS on Information and Systems...

    2020

  24. [32]

    Vimal Mollyn, Riku Arakawa, Mayank Goel, Chris Harrison, and Karan Ahuja

  25. [33]

    Bilge Mutlu, Andreas Krause, Jodi Forlizzi, Carlos Guestrin, and Jessica Hodgins

  26. [34]

    Muhammad Nadeem, Ersin Elbasi, Aymen I Zreikat, and Mohammad Sharsheer

  27. [35]

    Kento Nishimura, Kodai Ito, Ken Fujiwara, Kazuyuki Fujita, and Yuichi Itoh

  28. [36]

    Ali Nouriani, RA McGovern, and R Rajamani. 2024. Vector-based Inertial Poser: Human pose estimation with high gain observer and deep learning using sparse IMU sensors. Biomedical Signal Processing and Control 95 (2024), 106432

    2024

  29. [37]

    Marcus Ottosson. 2024. Ragdoll Dynamics. https://ragdolldynamics.com/. Ac- cessed: 2025-04-09

    2024

  30. [38]

    Varun Ramani, Hossein Khayami, Yang Bai, Nakul Garg, and Nirupam Roy. 2024. IMUOptimize: A Data-Driven Approach to Optimal IMU Placement for Human Pose Estimation with Transformer Architecture. arXiv preprint arXiv:2402.08923 (2024)

    2024 arXiv

  31. [39]

    Lala Shakti Swarup Ray, Daniel Geißler, Mengxi Liu, Bo Zhou, Sungho Suh, and Paul Lukowicz. 2024. ALS-HAR: Harnessing Wearable Ambient Light Sensors to Enhance IMU-Based Human Activity Recognition. In International Conference on Pattern Recognition. Springer, 133–147

    2024

  32. [40]

    Lala Shakti Swarup Ray, Lars Krupp, Vitor Fortes Rey, Bo Zhou, Sungho Suh, and Paul Lukowicz. 2025. TxP: Reciprocal Generation of Ground Pressure Dynamics and Activity Descriptions for Improving Human Activity Recognition. Proceed- ings of the ACM on Interactive, Mobile, Weara...

    2025

  33. [41]

    Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, Lars Krupp, Vitor Fortes Rey, and Paul Lukowicz. 2024. Text me the data: Generating ground pressure sequence from textual descriptions for har. In 2024 IEEE International Conference on Per- vasive Computing and Communications Worksh...

    2024

  34. [42]

    Quality and User Experience 8, 1 (2023), 10

    Detection of nodding of interlocutors using a chair-shaped device and investigating relationship between a divergent thinking task and amount of nodding. Quality and User Experience 8, 1 (2023), 10

    2023

  35. [43]

    Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, and Paul Lukowicz. 2023. Selecting the motion ground truth for loose-fitting wearables: Benchmarking optical mocap methods. In Proceedings of the 2023 ACM International Symposium on Wearable Computers. 27–32

    2023

  36. [44]

    Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, and Paul Lukowicz. 2024. A comprehensive evaluation of marker-based, markerless methods for loose gar- ment scenarios in varying camera configurations. Frontiers in Computer Science 6 (2024), 1379925

    2024

  37. [45]

    Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, and Paul Lukowicz. 2025. OV- HHIR: Open Vocabulary Human Interaction Recognition Using Cross-modal Integration of Large Language Models. In ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing ...

    2025

  38. [46]

    Narumon Roynarin, Sirinant Channak, and Prawit Janwantanakul. 2024. Postural shifts and body perceived discomfort during 1-hour sitting when leaning and sitting on an air-filled seat cushion among healthy office workers. Ergonomics 67, 12 (2024), 2126–2137

    2024

  39. [47]

    Jesse Scott, Bharadwaj Ravichandran, Christopher Funk, Robert T Collins, and Yanxi Liu. 2020. From image to stability: Learning dynamics from human pose. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIII 16 . Sprin...

    2020

  40. [48]

    Minwoo Seong, Gwangbin Kim, Jaehee Lee, Joseph DelPreto, Wojciech Matusik, Daniela Rus, and SeungJun Kim. 2024. Intelligent Seat: Tactile Signal-Based 3D Sitting Pose Inference. In Companion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computin...

    2024

  41. [49]

    Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, and Paul Lukowicz. 2023. Pressim: An end-to-end framework for dynamic ground pressure profile generation from monocular videos using physics-based 3d simulation. In 2023 IEEE International Conference on Pervasive Computing and Commu...

    2023

  42. [50]

    Soyong Shin, Juyong Kim, Eni Halilaj, and Michael J Black. 2024. Wham: Recon- structing world-grounded humans with accurate 3d motion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 2070–2080

    2024

  43. [51]

    Amritanshu Kumar Singh, Vedant Arvind Kumbhare, and K Arthi. 2021. Real- time human pose detection and recognition using mediapipe. In International conference on soft computing and signal processing . Springer, 145–154

    2021

  44. [52]

    Li Siyao, Weijiang Yu, Tianpei Gu, Chunze Lin, Quan Wang, Chen Qian, Chen Change Loy, and Ziwei Liu. 2022. Bailando: 3d dance generation by actor- critic gpt with choreographic memory. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 11050–11059

    2022

  45. [53]

    Adane Nega Tarekegn, Mohib Ullah, Faouzi Alaya Cheikh, and Muhammad Sajjad

  46. [54]

    Ming-Chih Tsai, Edward T-H Chu, and Chia-Rong Lee. 2023. An automated sitting posture recognition system utilizing pressure sensors. Sensors 23, 13 (2023), 5894

    2023

  47. [55]

    Patrick Vermander, Aitziber Mancisidor, Itziar Cabanes, and Nerea Perez. 2024. Intelligent systems for sitting posture monitoring and anomaly detection: an overview. Journal of neuroengineering and rehabilitation 21, 1 (2024), 28

    2024

  48. [56]

    Soshi Shimada, Vladislav Golyanik, Weipeng Xu, and Christian Theobalt. 2020. Physcap: Physically plausible monocular 3d motion capture in real time. ACM Transactions on Graphics (ToG) 39, 6 (2020), 1–16

    2020

  49. [57]

    Heng Wang, Jieqing Zheng, Qin Nie, Wenbo Li, Zhuo Wang, Kun Xiao, Xuehao Hu, Santosh Kumar, and Rui Min. 2024. Double-Fishtail-Shaped FBG wearable device for sitting posture recognition and real-time respiratory monitoring. IEEE Sensors Journal (2024)

    2024

  50. [58]

    Erwin Wu, Rawal Khirodkar, Hideki Koike, and Kris Kitani. 2024. SolePoser: Full Body Pose Estimation using a Single Pair of Insole Sensor. In Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology . 1–9

    2024

  51. [59]

    Ziyu Wu, Fangting Xie, Yiran Fang, Zhen Liang, Quan Wan, Yufan Xiong, and Xiaohui Cai. 2024. Seeing through the tactile: 3d human shape estimation from temporal in-bed pressure images. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 8, 2 (20...

    2024

  52. [60]

    Vasco Xu, Chenfeng Gao, Henry Hoffmann, and Karan Ahuja. 2024. Mobileposer: Real-time full-body pose estimation and 3d human translation from imus in mobile consumer devices. In Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology . 1–11

    2024

  53. [61]

    In 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)

    Enhancing human activity recognition through sensor fusion and hybrid deep learning model. In 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) . IEEE, 1–5

    2023

  54. [62]

    Tianhong Catherine Yu, Manru Mary Zhang, Peter He, Chi-Jung Lee, Cassidy Cheesman, Saif Mahmud, Ruidong Zhang, François Guimbretière, and Cheng Zhang. 2024. SeamPose: Repurposing Seams as Capacitive Sensors in a Shirt for Upper-Body Pose Tracking. In Proceedings of the 37th An...

    2024

  55. [63]

    Yu Zhang, Songpengcheng Xia, Lei Chu, Jiarui Yang, Qi Wu, and Ling Pei. 2024. Dynamic inertial poser (dynaip): Part-based motion dynamics learning for en- hanced human pose estimation with sparse inertial sensors. In Proceedings of the IEEE/CVF Conference on Computer Vision an...

    2024

  56. [64]

    Birgit Wallmann-Sperlich, Jens Bucksch, Sylvia Hansen, Peter Schantz, and Ingo Froboese. 2013. Sitting time in Germany: an analysis of socio-demographic and environmental correlates. BMC public health 13 (2013), 1–10

    2013

  57. [65]

    Weibing Zhong, Hui Xu, Yiming Ke, Xiaojuan Ming, Haiqing Jiang, Mufang Li, and Dong Wang. 2024. Accurate and Efficient Sitting Posture Recognition and Human-Machine Interaction Device Based on Fabric Pressure Sensor Array and Neural Network. Advanced Materials Technologies 9, ...

    2024

  58. [66]

    Bo Zhou, Daniel Geissler, Marc Faulhaber, Clara Elisabeth Gleiss, Es- ther Friederike Zahn, Lala Shakti Swarup Ray, David Gamarra, Vitor Fortes Rey, Sungho Suh, Sizhen Bian, et al . 2023. Mocapose: Motion capturing with textile-integrated capacitive sensors in loose-fitting sm...

    2023

  59. [67]

    Wentao Zhu, Xiaoxuan Ma, Zhaoyang Liu, Libin Liu, Wayne Wu, and Yizhou Wang. 2023. Motionbert: A unified perspective on learning human motion repre- sentations. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 15085–15099

    2023

  60. [68]

    Haolin Zhuang, Shun Lei, Long Xiao, Weiqin Li, Liyang Chen, Sicheng Yang, Zhiyong Wu, Shiyin Kang, and Helen Meng. 2023. Gtn-bailando: Genre consistent long-term 3d dance generation based on pre-trained genre token network. In ICASSP 2023-2023 IEEE International Conference on ...

    2023

  61. [69]

    Song Yan and Joni-Kristian Kämäräinen. 2021. Learning anthropometry from rendered humans. arXiv preprint arXiv:2101.02515 (2021)

    2021 arXiv

  62. [72]

    Mingjie Zhao, Fangting Xie, Ziyu Wu, Zhen Liang, and Xiaohui Cai. 2024. 3D Human Pose Estimation Using Pressure Images on a Smart Chair. In Proceedings of the 2024 2nd Asia Conference on Computer Vision, Image Processing and Pattern Recognition. 1–7

    2024

  63. [2007]

    In Proceedings of the 20th annual ACM symposium on User interface software and technology

    Robust, low-cost, non-intrusive sensing and recognition of seated postures. In Proceedings of the 20th annual ACM symposium on User interface software and technology. 149–158

  64. [2021]

    Electronics 10, 21 (2021), 2569

    Low-cost wireless wearable system for posture monitoring. Electronics 10, 21 (2021), 2569

    2021

  65. [2022]

    InProceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Generating diverse and natural 3d human motions from text. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition . 5152–5161

  66. [2023]

    In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems

    Imuposer: Full-body pose estimation using imus in phones, watches, and earbuds. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. 1–12

    2023

  67. [2024]

    Applied Sciences 14, 18 (2024), 8557

    Sitting posture recognition systems: Comprehensive literature review and analysis. Applied Sciences 14, 18 (2024), 8557

    2024

  68. [2025]

    IEEE Sensors Journal (2025)

    SitPose: Real-Time Detection of Sitting Posture and Sedentary Behavior Using Ensemble Learning With Depth Sensor. IEEE Sensors Journal (2025)

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.