Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose
Not on the map. Why?
Abstract (arXiv)
In this work we adapt multi-person pose estimation architecture to use it on edge devices. We follow the bottom-up approach from OpenPose, the winner of COCO 2016 Keypoints Challenge, because of its decent quality and robustness to number of people inside the frame. With proposed network design and optimized post-processing code the full solution runs at 28 frames per second (fps) on IntelunicodexAE NUC 6i7KYB mini PC and 26 fps on CoreTM i7-6850K CPU. The network model has 4.1M parameters and 9 billions floating-point operations (GFLOPs) complexity, which is just ~15% of the baseline 2-stage OpenPose with almost the same quality. The code and model are available as a part of IntelunicodexAE OpenVINOTM Toolkit.
Key facts. Cited by 42 papers in the map (data through 7 Oct 2026). First posted on arXiv 29 Nov 2018.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 13; 2022: 11; 2023: 10; 2024: 6; 2025: 0; 2026: 2.
42 papers.
Sort:
- 2021FrankMocap: A Monocular 3D Whole-Body Pose Estimation System via Regression and Integrationcited by 168
- 2021Online Knowledge Distillation for Efficient Pose Estimationcited by 55
- 2022DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimationcited by 29
- 2022PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimationcited by 27
- 2021A Lightweight Graph Transformer Network for Human Mesh Reconstruction from 2D Human Posecited by 21
- 2023Vision-Based Human Pose Estimation via Deep Learning: A Surveycited by 20
- 2022Lite Pose: Efficient Architecture Design for 2D Human Pose Estimationcited by 16
- 2024Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenescited by 11
- 2023PresSim: An End-to-end Framework for Dynamic Ground Pressure Profile Generation from Monocular Videos Using Physics-based 3D Simulationcited by 9
- 2022Fast and Robust Video-Based Exercise Classification via Body Pose Tracking and Scalable Multivariate Time Series Classifierscited by 8
- 2024GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimationcited by 6
- 2021Towards Fast and Accurate Multi-Person Pose Estimation on Mobile Devicescited by 5
30 more
- 2023Human Body Pose Estimation for Gait Identification: A Comprehensive Survey of Datasets and Modelscited by 5
- 2023VSViG: Real-time Video-based Seizure Detection via Skeleton-based Spatiotemporal ViGcited by 5
- 2022Multitask Network for Joint Object Detection, Semantic Segmentation and Human Pose Estimation in Vehicle Occupancy Monitoringcited by 4
- 2022Occupancy Planes for Single-view RGB-D Human Reconstructioncited by 4
- 2024DIR-BHRNet: A Lightweight Network for Real-time Vision-based Multi-person Pose Estimation on Smartphonescited by 4
- 2024Your Turn: At Home Turning Angle Estimation for Parkinson's Disease Severity Assessmentcited by 3
- 2021Towards Automated and Marker-less Parkinson Disease Assessment: Predicting UPDRS Scores using Sit-stand videoscited by 2
- 2021ARShoe: Real-Time Augmented Reality Shoe Try-on System on Smartphonescited by 2
- 2022ADG-Pose: Automated Dataset Generation for Real-World Human Pose Estimationcited by 2
- 2022Binarizing by Classification: Is soft function really necessary?cited by 2
- 2023Real-time Addressee Estimation: Deployment of a Deep-Learning Model on the iCub Robotcited by 2
- 2021Orderly Dual-Teacher Knowledge Distillation for Lightweight Human Pose Estimationcited by 1
- 2021MotionInput v2.0 supporting DirectX: A modular library of open-source gesture-based machine learning and computer vision methods for interacting and controlling existing software with a webcamcited by 1
- 2022MARIO: Modular and Extensible Architecture for Computing Visual Statistics in RoboCup SPLcited by 1
- 2023A Video-based Detector for Suspicious Activity in Examination with OpenPosecited by 1
- 2023A webcam-based machine learning approach for three-dimensional range of motion evaluationcited by 1
- 2023Efficient, Self-Supervised Human Pose Estimation with Inductive Prior Tuningcited by 1
- 2021A Global to Local Double Embedding Method for Multi-person Pose Estimationcited by 0
- 2021Efficient Realistic Data Generation Framework leveraging Deep Learning-based Human Digitizationcited by 0
- 2021Multi-label Classification of Aircraft Heading Changes Using Neural Network to Resolve Conflictscited by 0
- 2021A Review on Human Pose Estimationcited by 0
- 2021Rethinking Deconvolution for 2D Human Pose Estimation Light yet Accurate Model for Real-time Edge Computingcited by 0
- 2022DF-Captcha: A Deepfake Captcha for Preventing Fake Callscited by 0
- 2022InFi: End-to-End Learning to Filter Input for Resource-Efficiency in Mobile-Centric Inferencecited by 0
- 2023P=uioio: On-device Real-Time Smartphone-Based Automated Exercise Repetition Counting Systemcited by 0
- 2023Improved TokenPose with Sparsitycited by 0
- 2024Estimating Human Poses Across Datasets: A Unified Skeleton and Multi-Teacher Distillation Approachcited by 0
- 2024A Multi-Modal Explainability Approach for Human-Aware Robots in Multi-Party Conversationcited by 0
- 2026Variational Encoder--Multi-Decoder (VE-MD) for Privacy-by-functional-design (Group) Emotion Recognitioncited by 0
- 2026Unsupervised Keypoints for Real-Time Fall Detection: Comparative Analysis Under Real-world Conditions with Predictive Bandwidth Reductioncited by 0
- 2026Unsupervised Keypoints for Real-Time Fall Detection: Comparative Analysis Under Real-world Conditions with Predictive Bandwidth Reduction16 Jul 2026
- 2026Variational Encoder--Multi-Decoder (VE-MD) for Privacy-by-functional-design (Group) Emotion Recognition2 Apr 2026
- 2024Your Turn: At Home Turning Angle Estimation for Parkinson's Disease Severity Assessment15 Aug 2024
- 2024GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimation15 Jul 2024
- 2024DIR-BHRNet: A Lightweight Network for Real-time Vision-based Multi-person Pose Estimation on Smartphones1 Jul 2024
- 2024Estimating Human Poses Across Datasets: A Unified Skeleton and Multi-Teacher Distillation Approach30 May 2024
- 2024A Multi-Modal Explainability Approach for Human-Aware Robots in Multi-Party Conversation20 May 2024
- 2024Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes27 Mar 2024
- 2023VSViG: Real-time Video-based Seizure Detection via Skeleton-based Spatiotemporal ViG24 Nov 2023
- 2023Improved TokenPose with Sparsity16 Nov 2023
- 2023Real-time Addressee Estimation: Deployment of a Deep-Learning Model on the iCub Robot9 Nov 2023
- 2023Efficient, Self-Supervised Human Pose Estimation with Inductive Prior Tuning6 Nov 2023
30 more
- 2023A webcam-based machine learning approach for three-dimensional range of motion evaluation11 Oct 2023
- 2023Vision-Based Human Pose Estimation via Deep Learning: A Survey26 Aug 2023
- 2023P=uioio: On-device Real-Time Smartphone-Based Automated Exercise Repetition Counting System22 Jul 2023
- 2023A Video-based Detector for Suspicious Activity in Examination with OpenPose21 Jul 2023
- 2023Human Body Pose Estimation for Gait Identification: A Comprehensive Survey of Datasets and Models23 May 2023
- 2023PresSim: An End-to-end Framework for Dynamic Ground Pressure Profile Generation from Monocular Videos Using Physics-based 3D Simulation1 Feb 2023
- 2022Fast and Robust Video-Based Exercise Classification via Body Pose Tracking and Scalable Multivariate Time Series Classifiers2 Oct 2022
- 2022InFi: End-to-End Learning to Filter Input for Resource-Efficiency in Mobile-Centric Inference28 Sep 2022
- 2022MARIO: Modular and Extensible Architecture for Computing Visual Statistics in RoboCup SPL20 Sep 2022
- 2022PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimation16 Sep 2022
- 2022DF-Captcha: A Deepfake Captcha for Preventing Fake Calls17 Aug 2022
- 2022Occupancy Planes for Single-view RGB-D Human Reconstruction4 Aug 2022
- 2022Binarizing by Classification: Is soft function really necessary?16 May 2022
- 2022Multitask Network for Joint Object Detection, Semantic Segmentation and Human Pose Estimation in Vehicle Occupancy Monitoring3 May 2022
- 2022Lite Pose: Efficient Architecture Design for 2D Human Pose Estimation3 May 2022
- 2022DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation16 Mar 2022
- 2022ADG-Pose: Automated Dataset Generation for Real-World Human Pose Estimation1 Feb 2022
- 2021A Lightweight Graph Transformer Network for Human Mesh Reconstruction from 2D Human Pose24 Nov 2021
- 2021Rethinking Deconvolution for 2D Human Pose Estimation Light yet Accurate Model for Real-time Edge Computing8 Nov 2021
- 2021A Review on Human Pose Estimation13 Oct 2021
- 2021Multi-label Classification of Aircraft Heading Changes Using Neural Network to Resolve Conflicts10 Sep 2021
- 2021ARShoe: Real-Time Augmented Reality Shoe Try-on System on Smartphones24 Aug 2021
- 2021FrankMocap: A Monocular 3D Whole-Body Pose Estimation System via Regression and Integration13 Aug 2021
- 2021MotionInput v2.0 supporting DirectX: A modular library of open-source gesture-based machine learning and computer vision methods for interacting and controlling existing software with a webcam10 Aug 2021
- 2021Online Knowledge Distillation for Efficient Pose Estimation4 Aug 2021
- 2021Efficient Realistic Data Generation Framework leveraging Deep Learning-based Human Digitization28 Jun 2021
- 2021Towards Fast and Accurate Multi-Person Pose Estimation on Mobile Devices6 Jun 2021
- 2021Orderly Dual-Teacher Knowledge Distillation for Lightweight Human Pose Estimation21 Apr 2021
- 2021Towards Automated and Marker-less Parkinson Disease Assessment: Predicting UPDRS Scores using Sit-stand videos10 Apr 2021
- 2021A Global to Local Double Embedding Method for Multi-person Pose Estimation15 Feb 2021
- 2021FrankMocap: A Monocular 3D Whole-Body Pose Estimation System via Regression and Integrationcited by 168
- 2021Online Knowledge Distillation for Efficient Pose Estimationcited by 55
- 2022DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimationcited by 29
- 2022PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimationcited by 27
- 2021A Lightweight Graph Transformer Network for Human Mesh Reconstruction from 2D Human Posecited by 21
- 2023Vision-Based Human Pose Estimation via Deep Learning: A Surveycited by 20
- 2022Lite Pose: Efficient Architecture Design for 2D Human Pose Estimationcited by 16
- 2024Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenescited by 11
- 2023PresSim: An End-to-end Framework for Dynamic Ground Pressure Profile Generation from Monocular Videos Using Physics-based 3D Simulationcited by 9
- 2022Fast and Robust Video-Based Exercise Classification via Body Pose Tracking and Scalable Multivariate Time Series Classifierscited by 8
- 2024GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimationcited by 6
- 2021Towards Fast and Accurate Multi-Person Pose Estimation on Mobile Devicescited by 5
12 most cited of 42 All 42 (JSON)
- 2026Unsupervised Keypoints for Real-Time Fall Detection: Comparative Analysis Under Real-world Conditions with Predictive Bandwidth Reduction16 Jul 2026
- 2026Variational Encoder--Multi-Decoder (VE-MD) for Privacy-by-functional-design (Group) Emotion Recognition2 Apr 2026
- 2024Your Turn: At Home Turning Angle Estimation for Parkinson's Disease Severity Assessment15 Aug 2024
- 2024GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimation15 Jul 2024
- 2024DIR-BHRNet: A Lightweight Network for Real-time Vision-based Multi-person Pose Estimation on Smartphones1 Jul 2024
- 2024Estimating Human Poses Across Datasets: A Unified Skeleton and Multi-Teacher Distillation Approach30 May 2024
- 2024A Multi-Modal Explainability Approach for Human-Aware Robots in Multi-Party Conversation20 May 2024
- 2024Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes27 Mar 2024
- 2023VSViG: Real-time Video-based Seizure Detection via Skeleton-based Spatiotemporal ViG24 Nov 2023
- 2023Improved TokenPose with Sparsity16 Nov 2023
- 2023Real-time Addressee Estimation: Deployment of a Deep-Learning Model on the iCub Robot9 Nov 2023
- 2023Efficient, Self-Supervised Human Pose Estimation with Inductive Prior Tuning6 Nov 2023
12 newest of 42 All 42 (JSON)