Pith. sign in

REVIEW 4 major objections 4 minor 27 references

PosePilot: An Edge-AI Solution for Posture Correction in Physical Exercises

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read PosePilot treats posture correction as continuously forecasting the next joint-angle vector, flagging deviations at every step of a yoga pose.

desk verdict Plausible recognition pipeline plus a genuinely new dataset, but the pose-correction claim rests on forecasting MSE and an unverified assumption that the training videos are correct executions. read the letter →

arxiv 2505.19186 v1 pith:H6JPBP2Z submitted 2025-05-25 cs.CV cs.AI

Rushiraj Gadhvi , Priyansh Desai , Siddharth This is my paper · ORCID
classification cs.CVcs.AI
keywords yogaposecorrectionrecognitionBiLSTMmulti-headattentionedgeAIjointanglescorrectivefeedbacktemporalforecasting
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

The paper sets out to show that automated posture correction can be recast as trajectory forecasting rather than final-pose classification. Using Yoga as the test case, a two-layer bidirectional LSTM learns the joint-angle history of correctly performed asanas and predicts the next angle vector, so any angle that strays more than 1.5 standard deviations from the forecast is flagged with a direction to adjust. This gives feedback continuously through a movement rather than only at the end, which the paper identifies as the missing piece in prior correction systems. If the claim holds, users can receive personalized, on-device coaching without internet or cloud processing.

What carries the argument

The load-bearing object is the forecasted joint-angle vector $\hat{p}_t$ for nine body angles: left and right shoulders, elbows, hips, knees, and the neck. The two-layer BiLSTM consumes standardized sequences of these angles and predicts the next vector; the error module compares the practitioner's actual angles with $\hat{p}_t$ and flags any angle exceeding 1.5 standard deviations, using the vector difference to show the direction of the needed adjustment. This converts correction into a next-frame forecasting task, which is what enables per-instant feedback rather than a single end-of-pose verdict.

What would settle it

Have yoga instructors independently annotate the frames where deliberately incorrect asanas deviate from correct form, then compare their labels with the model's 1.5-standard-deviation flags; if the flags and expert labels do not overlap substantially, the correction claim is not supported.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that pose correction becomes a sequence-forecasting problem: a two-layer BiLSTM with multi-head attention is trained on the temporal pattern of nine joint angles sampled from correctly performed asanas, and at each time step it forecasts the next joint-angle vector. Any of the user's nine angles that deviates by more than 1.5 standard deviations from that forecast is reported as an error requiring adjustment, with the direction of correction coming from the vector difference. Because the forecast is produced at every frame, the feedback applies to intermediate sub-poses, not just the final posture. The recognition module classifies the ongoing asana and selects one of six per-pose correction models. On an in-house dataset of 336 videos (14 participants, six asanas, four camera angles), the paper reports 97.52% recognition accuracy and average correction-model MSE of 0.00138; quantized to 8-bit integers on a Raspberry Pi 4, recognition runs at 330.65 FPS and correction at 6.42 FPS, with reported accuracy losses of about 1.0-1.3%.

Load-bearing premise

The load-bearing premise is that the in-house videos are faithful executions of each asana, yet the paper does not report expert verification of the 14 participants' performances; if those recordings contain imperfect form, the forecast guides users toward the dataset's average movement rather than a verified correct posture.

Editorial extensions

If this is right

  • Users receive correction signals throughout a movement: each frame that deviates from the forecasted angle trajectory is flagged, so intermediate sub-poses are corrected rather than only the final pose.
  • Feedback is angle-specific and directional: the system identifies which of the nine monitored body angles is off and in which direction to adjust, instead of merely classifying the pose as incorrect.
  • The recognition-plus-correction loop is designed to run on a Raspberry Pi-class edge device at usable frame rates, which would make personalized coaching available offline and keep exercise video local.
  • Because the correction model is selected per recognized asana, the same architecture can be retrained for other sequence-based physical activities named in the paper, such as rehabilitation and sports coaching.

Reading between the lines

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

  • An implicit test the paper leaves open is whether the 1.5-standard-deviation flagging rule agrees with expert judgment: comparing system flags on deliberately incorrect asanas with annotations from yoga instructors would settle whether the threshold marks the same mistakes a human coach would.
  • A corollary not developed in the paper is that errors in the recognition stage would route the user to the wrong correction model, so end-to-end correctness depends on recognition confidence and would benefit from an explicit uncertainty estimate.
  • The statistical notion of 'correct' here is the center of the training distribution; with a larger, expert-validated, body-diverse dataset, the learned trajectory could become a coaching-grounded standard rather than only the average of the recorded movements.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. PosePilot proposes an edge-AI system for yoga pose recognition and corrective feedback. It introduces an in-house multi-angle video dataset of six asanas from 14 participants, extracts 680 joint angles via MediaPipe, uses a key-frame selection heuristic and an LSTM with attention for classification (reported accuracy 97.52%, F1 0.99), and a two-layer BiLSTM with attention that forecasts the next nine-angle vector; deviations above 1.5 standard deviations are flagged as errors and displayed as corrective feedback. The authors also report deployment on a Raspberry Pi 4 with INT8 quantization. The main claimed contribution is 'instant corrective feedback at every stage of a movement,' and the paper argues that prior systems only address final-pose correction rather than sub-pose feedback.

Significance. If the correction claim were fully validated, PosePilot would be a useful contribution: it addresses an underexplored problem (sub-pose correction, not just final-pose classification), reports a new multi-angle yoga dataset, and includes an edge-deployment study with a code repository link. The recognition component is reasonably evaluated with 10-fold cross-validation and reports a high F1 score. The paper's strength is the integration of angle-based LSTM recognition, forecasting-based correction, and edge quantization. However, the central correction claim is not yet established by the evidence presented: the ideal-pose ground truth is internal to the training set, and no detection-level metrics are reported. The significance is therefore conditional on the additional validation requested below.

major comments (4)
  1. [Section 3.1 and 3.3] The correction model's core premise is that the BiLSTM forecast represents the correct/ideal trajectory, but the paper never establishes that the in-house training videos are expert-validated correct executions. Section 3.1 criticizes prior datasets for using non-professional practitioners, yet it only states that the 14 participants were recorded, not that their performances were checked by a yoga expert; the only expert involvement reported (Section 3.3) is in selecting the nine angles for correction. If the training clips contain imperfect or non-representative executions, the 1.5-standard-deviation flags steer users toward the dataset's average movement rather than a verified safe posture. Please state explicitly whether expert validation was performed; if not, either add it or re-frame the system as deviation-from-learned-pattern feedback and provide external validation.
  2. [Section 4.2] The evaluation of the correction model does not test the correction claim. The only quantitative metric is forecasting MSE of 0.00138 on a held-out split (Table 2), which measures the model's ability to predict the next angle within the same dataset distribution; it does not measure whether a flagged deviation corresponds to a real postural error. There is no precision/recall against expert-labeled error frames, no comparison with a baseline forecaster (e.g., persistence or linear extrapolation), and the 'deliberately incorrect asanas' experiment is presented only as a qualitative figure without detection accuracy. The MSE is also not interpretable without the angle scale or a baseline. Please add a detection-level evaluation with expert-labeled errors, report precision/recall or F1 for the error flag, and compare the forecaster against a simple baseline.
  3. [Section 3.2] The key-frame hyperparameter k is selected on test data: the text states 'During testing, we found that k=10 yielded optimal results.' This makes the reported recognition accuracy potentially optimistic and the selection procedure non-standard. In addition, the description of data augmentation does not clarify whether augmentation is applied before or after the train/test split; if before, augmented copies of test sequences can leak into training. Please use a validation split or nested cross-validation for hyperparameter selection and describe the augmentation/split order explicitly.
  4. [Section 4.3] The real-time claim is only partially supported: the pose-correction model alone runs at 6.42 FPS after INT8 quantization, but feature extraction on video frames is excluded and is stated to need further optimization. The end-to-end latency (camera capture + MediaPipe keypoint extraction + angle computation + recognition + correction + GUI rendering) is not reported. Since the abstract promises 'instant corrective feedback at every stage of a movement,' please provide an end-to-end frame-rate or latency measurement on the deployed hardware, or soften the real-time claim accordingly.
minor comments (4)
  1. [Table 1] In Table 1, the first two rows both list 'Yadav et al. [27]' as the dataset, which appears to be a copy-paste error; the second row should cite the Infinity Yoga Tutor entry.
  2. [Equations 2a-2c] Equation (2a) uses the symbol p both as the interval length and in the summation bounds; please clarify the window size (e.g., p=5 for t-2 to t+2) and define the index range unambiguously.
  3. [Section 3.3] The correction target range is described as 'within one standard deviation from the predicted value,' but it is not stated whether the standard deviation is computed per angle, per pose, or across the dataset; please specify.
  4. [Section 4.1] The reported F1 score of 0.99 and accuracy of 97.52% are not accompanied by confidence intervals or per-class breakdowns; please include these to support the recognition claim.

Circularity Check

2 steps flagged · score 6.0 of 10

Pose correction's 'ideal pose' is the model's own BiLSTM forecast; the only quantitative evaluation is in-sample forecasting MSE, so the central correction claim is partially circular.

  1. self definitional [Section 3.3 and Figure 3 caption]
    "If the difference between the predicted and actual angle values exceeds 1.5 standard deviations, the point is flagged as an error, indicating a potential deviation from expected performance requiring correction. ... Any joint angle deviating more than 1.5 standard deviations from the ideal pose is flagged, and red vectors show the adjustment needed to bring each point back within the acceptable range."

    The 'ideal pose' used for correction is the BiLSTM's one-step forecast, and that forecast is trained on the in-house videos. Section 3.1 criticizes prior datasets because practitioners are 'generally not professionals, thus limiting the establishment of the ground truth baseline for pose correction,' but the paper never states that its own 14 participants were expert-validated. Therefore 'correct posture' is defined as whatever the fitted model predicts; feedback flags deviations from the model's own learned trajectory rather than from an external expert standard. By construction, ideal pose = model prediction, so the correction output is self-referential.

  2. fitted input called prediction [Section 4.2]
    "The performance of the model was evaluated using a forecasting task on the in-house test dataset, where it achieved an average mean square error (MSE) of 0.00138. ... This low MSE indicates that the model was highly accurate in predicting correct pose trajectories. As no external benchmarks are available (as detailed in Section 3.1), we therefore relied solely on our test dataset for evaluation."

    The reported MSE is a forecasting fit on a held-out split of the same unvalidated in-house videos; it measures how well the BiLSTM reproduces the dataset's average trajectory, not whether a flagged deviation corresponds to an actual postural error. Calling this 'predicting correct pose trajectories' renames fit quality as correction accuracy. The 'deliberately incorrect asanas' experiment also defines incorrectness by the same 1.5-standard-deviation rule, so it can only confirm the model's internal threshold, not the correctness of the feedback. The correction claim's only quantitative evidence is the model's agreement with its own training distribution.

full rationale

PosePilot's correction mechanism is internally consistent but not externally grounded. The paper's own equations define an error as a deviation from the BiLSTM's predicted next joint-angle vector, and the Figure 3 caption equates that prediction with 'the ideal pose'. Since the BiLSTM is trained on in-house videos whose correctness is asserted but never independently validated, the standard of correctness is internal to the model. The evaluation section does not break this circle: the 0.00138 MSE is a forecasting metric on a test split of the same dataset, and the deliberately incorrect asana experiment relies on the same 1.5-sigma threshold to label errors, so it tests self-consistency rather than real postural correctness. Pose recognition, by contrast, is evaluated against an external dataset (Yadav et al.) and has independent content. Weighing these, the central novel claim of personalized corrective feedback reduces partly to the model's fitted trajectory, warranting a score of 6 rather than a higher score, because the recognition component and edge-deployment evaluation are not circular.

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

The paper builds on standard machine learning components but introduces several hand-set hyperparameters and implicit assumptions. The most critical is that unvalidated participant videos define ground-truth correctness for the correction model; this is not an external benchmark. No new physical entities are introduced.

free parameters (5)
  • Number of key frames k = 10
    Chosen during testing as optimal for recognition; affects input length of LSTM and reported accuracy.
  • Temporal window p for deviation = 5 frames (t-2 to t+2)
    Hand-selected window size in the aggregated standard deviation calculation for key frame selection and sequence normalization.
  • Error flag threshold = 1.5 standard deviations
    Arbitrary threshold used to flag a joint angle as incorrect relative to the BiLSTM forecast; no sensitivity analysis is provided.
  • Correction target range = within 1.0 standard deviation
    Feedback goal used in the GUI; chosen without evidence or validation.
  • Correction feature subset = 9 joint angles
    Reduced from 680 angles based on domain knowledge and yoga expert input; not validated by ablations.
assumptions (5)
  • domain assumption Mediapipe keypoint extraction provides accurate keypoints across the recorded viewpoints and lighting conditions.
    All downstream angle features and forecasts inherit the accuracy of the 33-keypoint pose estimator; no correction or uncertainty handling is provided in Section 3.1.
  • domain assumption The 14 participants performed each asana correctly, making the training videos a valid ground truth for correct posture.
    The correction model learns the ideal trajectory from these videos; the paper does not mention expert validation or professional status of the participants in Section 3.1.
  • domain assumption Deviation of a joint angle from the BiLSTM forecast is a valid measure of posture error.
    Error detection in Section 3.3 compares actual angles with predicted angles and flags values above 1.5 standard deviations; this assumes the forecast equals the correct trajectory and that angle differences are meaningful in the arccosine space.
  • domain assumption The 680 angles from all triplets of 17 keypoints are sufficient and redundant features for both recognition and correction.
    The feature set is constructed from all combinations, with no feature selection for recognition in Section 3.1; it also drives the key frame selection.
  • domain assumption Local maxima of aggregated angle standard deviation identify the most informative frames.
    Key frame selection in Section 3.2 relies on this hypothesis; no comparison with uniform frame sampling is reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PosePilot: An Edge-AI Solution for Posture Correction in Physical Exercises." pith.science (2026). https://pith.science/paper/H6JPBP2Z

@misc{pith2026250519186,
  author       = {Pith},
  title        = {Pith review of: PosePilot: An Edge-AI Solution for Posture Correction in Physical Exercises},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H6JPBP2Z}},
  note         = {Machine review of arXiv:2505.19186}
}
read the original abstract

Automated pose correction remains a significant challenge in AI-driven fitness systems, despite extensive research in activity recognition. This work presents PosePilot, a novel system that integrates pose recognition with real-time personalized corrective feedback, overcoming the limitations of traditional fitness solutions. Using Yoga, a discipline requiring precise spatio-temporal alignment as a case study, we demonstrate PosePilot's ability to analyze complex physical movements. Designed for deployment on edge devices, PosePilot can be extended to various at-home and outdoor exercises. We employ a Vanilla LSTM, allowing the system to capture temporal dependencies for pose recognition. Additionally, a BiLSTM with multi-head Attention enhances the model's ability to process motion contexts, selectively focusing on key limb angles for accurate error detection while maintaining computational efficiency. As part of this work, we introduce a high-quality video dataset used for evaluating our models. Most importantly, PosePilot provides instant corrective feedback at every stage of a movement, ensuring precise posture adjustments throughout the exercise routine. The proposed approach 1) performs automatic human posture recognition, 2) provides personalized posture correction feedback at each instant which is crucial in Yoga, and 3) offers a lightweight and robust posture correction model feasible for deploying on edge devices in real-world environments.

Figures

Figures reproduced from arXiv: 2505.19186 by the authors.

Figure 1
Figure 1. Sample of In-house Dataset. 3.2 Pose Recognition Model Our goal is to build a model that can classify Yoga asanas using sequential images in the video as input. However, since each pose sequence is recorded at practitioner’s own pace, the video sequences vary in length from person to person and contain a substantial amount of redundant information. To address this, we extract the key frames, a subset of frames that … view at source ↗
Figure 2
Figure 2. PosePilot Overview. Video frame’s extracted joint angles from time [0, t] are first fed into an LSTM that captures motion and classifies the current yoga pose, selecting one of six trained correction models. A BiLSTM then analyzes those frames to forecast the next joint-angle vector pˆt. These deviations are processed by an error-calculation module, and the user interface presents per-angle correction feedback along… view at source ↗
Figure 3
Figure 3. Correction graph for an incorrectly performed Utkatasana. Error corresponding frames are marked with red crosses. Any joint angle deviating more than 1.5 standard deviations from the ideal pose is flagged, and red vectors show the adjustment needed to bring each point back within the acceptable range. A correctly performed Utkatasana is shown alongside for comparison. range, defined as within one standard deviation … view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Pose Recognition and Personalized Corrective Feedback GUI complexity. The accuracy of pose recognition remained stable, with a variance of less than 2.5%. The pose correction model also demonstrated consistent feed￾back accuracy, with an MSE fluctuation within 0.0006 o…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 20 canonical work pages

  1. [1]

    SN Computer Science4(02 2023)

    Ashraf, F.B., Islam, M.U., Kabir, M.R., Uddin, J.: Yonet: A neural network for yoga pose classification. SN Computer Science4(02 2023). https://doi.org/10. 1007/s42979-022-01618-8

    work page 2023

  2. [2]

    Journal of Orthopaedics and Traumatology5, 188–190 (12 2004)

    Bianchi, G., Cavenago, C., Marchese, M.: Can the practice of yoga be dangerous? considerations over a case of epiphyseal separation of the distal tibia in a teenager. Journal of Orthopaedics and Traumatology5, 188–190 (12 2004). https://doi.org/ 10.1007/s10195-004-0069-y

    work page doi:10.1007/s10195-004-0069-y 2004

  3. [3]

    British Journal of Sports Medicine54, bjsports–2019 (08 2019)

    Bolling, C., Barboza, S., van Mechelen, W., Pasman, H.: Letting the cat out of the bag: Athletes, coaches and physiotherapists share their perspectives on injury prevention in elite sports. British Journal of Sports Medicine54, bjsports–2019 (08 2019). https://doi.org/10.1136/bjsports-2019-100773

    work page doi:10.1136/bjsports-2019-100773 2019

  4. [4]

    Chaudhari, A., Dalvi, O., Ramade, O., Ambawade, D.: Yog-guru: Real-time yoga pose correction system using deep learning methods. pp. 1–6 (2021). https://doi. org/10.1109/ICCICT50803.2021.9509937

    work page Pith review arXiv 2021

  5. [5]

    arXiv preprint 2006.11718 (2020)

    Chen, S., Yang, R.: Pose trainer: Correcting exercise posture using pose estimation. arXiv preprint 2006.11718 (2020). https://doi.org/10.48550/arXiv.2006.11718

    arXiv doi:10.48550/arxiv.2006.11718 2020

  6. [6]

    In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

    Dittakavi, B., Bavikadi, D., Desai, S.V., Chakraborty, S., Reddy, N., Balasub- ramanian, V.N., Callepalli, B., Sharma, A.: Pose tutor: An explainable system for pose correction in the wild. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). pp. 3539–3548 (2022). https://doi.org/10.1109/CVPRW56347.2022.00398

    arXiv 2022

  7. [7]

    In: IEEE/CVF Winter Conference on Applications of Computer Vision

    Dittakavi, B., Callepalli, B., Vardhan, A., Desai, S.V., Balasubramanian, V.N.: Care: Counterfactual-based algorithmic recourse for explainable pose correction. In: IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 4902– 4911 (2024). https://doi.org/10.1109/WACV57701.2024.00483

    arXiv 2024

  8. [8]

    Electronics11, 1785 (06 2022)

    Gad, G., Gad, E., Cengiz, K., Fadlullah, Z., Mokhtar, B.: Deep learning-based context-aware video content analysis on iot devices. Electronics11, 1785 (06 2022). https://doi.org/10.3390/electronics11111785

    work page doi:10.3390/electronics11111785 2022

Show all 27 references
  1. [9]

    IEEE Internet of Things Journal 6, 7192–7200 (2019)

    Gochoo, M., Tan, T.H., Yeh, w.c., Batjargal, T., Hsieh, J.W., Alnajjar, F., Chen, Y.F.: Novel iot-based privacy-preserving yoga posture recognition system using low-resolution infrared sensors and deep learning. IEEE Internet of Things Journal 6, 7192–7200 (2019). https://doi....

    2019

  2. [10]

    Neural Computing and Applications33, 1–15 (06 2021)

    Jain, S., Rustagi, A., Saurav, S., Saini, R., Singh, S.: Three-dimensional cnn- inspired deep learning architecture for yoga pose recognition in the real-world environment. Neural Computing and Applications33, 1–15 (06 2021). https: //doi.org/10.1007/s00521-020-05405-5

    2021 doi

  3. [11]

    Kelly, Philip and Healy, Aoife and Moran, Kieran and O’Connor, Noel E

    Kelly, P., Healy, A., Moran, K., O’Connor, N.: A virtual coaching environment for improving golf swing technique. Kelly, Philip and Healy, Aoife and Moran, Kieran and O’Connor, Noel E. (2010) A virtual coaching environment for improving golf swing technique. In: SMVC 2010 - AC...

    2010

  4. [12]

    International Journal of Industrial Ergonomics99, 1035–40 (2024)

    bin Kim, S., hyun Kim, S., bin Lim, O., hwi Yi, C., hyun Han, G.: Effects of a posture correction feedback system on neck and trunk posture and muscle activity during computer work. International Journal of Industrial Ergonomics99, 1035–40 (2024). https://doi.org/10.1016/j.ape...

    2024

  5. [13]

    Journal of sports sciences 20, 813–28 (11 2002)

    Lees, A.: Technique analysis in sports: A critical review. Journal of sports sciences 20, 813–28 (11 2002). https://doi.org/10.1080/026404102320675657 12 R. Gadhvi, P. Desai, Siddharth et al

    2002 doi

  6. [14]

    Educational Technology Research and Development69(01 2021)

    Lin, K.C., Wei, C.W., Lai, C.L., Cheng, I.L., Chen, N.S.: Development of a bad- minton teaching system with wearable technology for improving students’ bad- minton doubles skills. Educational Technology Research and Development69(01 2021). https://doi.org/10.1007/s11423-020-09935-6

    2021 doi

  7. [15]

    https://doi

    Lugaresi, C., Tang, J., Nash, H., McClanahan, C., Uboweja, E., Hays, M., Zhang, F., Chang, C.L., Yong, M., Lee, J., Chang, W.T., Hua, W., Georg, M., Grundmann, M.: Mediapipe: A framework for building perception pipelines (2019). https://doi. org/10.48550/arXiv.1906.08172

    2019 doi arXiv

  8. [16]

    British Journal of Sports Medicine54, bjsports–2018 (03 2020)

    Nielsen, R., Bertelsen, M., Moller, M., Hulme, A., Mansournia, m.a., Casals, M., Parner, E.: Methods matter: exploring the ‘too much, too soon’ theory, part 1: causal questions in sports injury research. British Journal of Sports Medicine54, bjsports–2018 (03 2020). https://do...

    2018 doi

  9. [17]

    Neural Computing and Applications28 (12 2017)

    Pai, P.F., ChangLiao, L.H., Lin, K.P.: Analyzing basketball games by a support vector machines with decision tree model. Neural Computing and Applications28 (12 2017). https://doi.org/10.1007/s00521-016-2321-9

    2017 doi

  10. [18]

    IEEE AccessPP, 1–1 (01 2023)

    Raza, A., Qadri, A., Akhtar, I., Abdelsamee, N., Alabdulhafith, M.: Logrf: An approach to human pose estimation using skeleton landmarks for physiotherapy fitness exercise correction. IEEE AccessPP, 1–1 (01 2023). https://doi.org/10. 1109/ACCESS.2023.3320144

    2023

  11. [19]

    In: 2020 5th International Conference on Information Technology Research (ICITR)

    Rishan,F.,DeSilva,B.,Alawathugoda,S.,Nijabdeen,S.,Rupasinghe,L.,Liyanap- athirana, C.: Infinity yoga tutor: Yoga posture detection and correction system. In: 2020 5th International Conference on Information Technology Research (ICITR). pp. 1–6 (2020). https://doi.org/10.1109/I...

    2020

  12. [20]

    In: 2015 10th Asian Control Conference (ASCC)

    Shan, C.Z., Ming, E.S.L., Rahman, H.A., Fai, Y.C.: Investigation of upper limb movement during badminton smash. In: 2015 10th Asian Control Conference (ASCC). pp. 1–6 (2015). https://doi.org/10.1109/ASCC.2015.7244605

    2015

  13. [21]

    https://doi.org/10.1109/BioRob.2012.6290668

    Taati, B., Wang, R., Huq, R., Snoek, J., Mihailidis, A.: Vision-based posture assess- ment to detect and categorize compensation during robotic rehabilitation therapy (2012). https://doi.org/10.1109/BioRob.2012.6290668

    2012

  14. [22]

    Complementary Therapies in Medicine57, 1026–44 (2020)

    Telles, S., Sharma, S., Chetry, D., Balkrishna, A.: Benefits and adverse effects associated with yoga practice: A cross-sectional survey from india. Complementary Therapies in Medicine57, 1026–44 (2020). https://doi.org/10.1016/j.ctim.2020. 102644

    2020 doi

  15. [23]

    In: 2019 International Confer- ence on Advanced Information Technologies (ICAIT)

    Thar, M.C., Winn, K.Z.N., Funabiki, N.: A proposal of yoga pose assessment method using pose detection for self-learning. In: 2019 International Confer- ence on Advanced Information Technologies (ICAIT). pp. 137–142 (2019). https: //doi.org/10.1109/AITC.2019.8920892

    2019

  16. [24]

    Verma, M., Kumawat, S., Nakashima, Y., Raman, S.: Yoga-82: A new dataset for fine-grained classification of human poses. pp. 1038–1039 (2020). https://doi.org/ 10.48550/arXiv.2004.10362

    2020 doi arXiv

  17. [25]

    Applied Sciences13, 13213 (12 2023)

    Vives, F., Lázaro, J., Guzmán, J., Martínez-Gallego, R., Crespo, M.: Optimizing sporting actions effectiveness: A machine learning approach to uncover key vari- ables in the men’s professional doubles tennis serve. Applied Sciences13, 13213 (12 2023). https://doi.org/10.3390/a...

    2023 doi

  18. [26]

    In: 2016 IEEE International Conference on Big Data (Big Data)

    Wang, J., Huang, Z., Zhang, W., Patil, A., Patil, K., Zhu, T., Shiroma, E.J., Schepps, M.A., Harris, T.B.: Wearable sensor based human posture recognition. In: 2016 IEEE International Conference on Big Data (Big Data). pp. 3432–3438 (2016). https://doi.org/10.1109/BigData.2016.7841004

    2016

  19. [27]

    Neural Computing and Applications31, 9349–9361 (2019)

    Yadav, S.K., Singh, A., Gupta, A., Raheja, J.L.: Real-time yoga recognition using deep learning. Neural Computing and Applications31, 9349–9361 (2019). https: //doi.org/10.1007/s00521-019-04232-7

    2019 doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.