Deep learning-based estimation of whole-body kinematics from multi-view images (2307.05896v1)
Abstract: It is necessary to analyze the whole-body kinematics (including joint locations and joint angles) to assess risks of fatal and musculoskeletal injuries in occupational tasks. Human pose estimation has gotten more attention in recent years as a method to minimize the errors in determining joint locations. However, the joint angles are not often estimated, nor is the quality of joint angle estimation assessed. In this paper, we presented an end-to-end approach on direct joint angle estimation from multi-view images. Our method leveraged the volumetric pose representation and mapped the rotation representation to a continuous space where each rotation was uniquely represented. We also presented a new kinematic dataset in the domain of residential roofing with a data processing pipeline to generate necessary annotations for the supervised training procedure on direct joint angle estimation. We achieved a mean angle error of $7.19\circ$ on the new Roofing dataset and $8.41\circ$ on the Human3.6M dataset, paving the way for employment of on-site kinematic analysis using multi-view images.
- BLS, 2019. National census of fatal occupational injuries in 2019. https://www.bls.gov/iif/oshwc/cfoi/cftb0331.htm.
- Twin gaussian processes for structured prediction. International Journal of Computer Vision 87, 28–52.
- Keep it smpl: Automatic estimation of 3d human pose and shape from a single image, in: ECCV.
- Human pose estimation with iterative error feedback. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , 4733–4742.
- Learnable human mesh triangulation for 3d human pose and shape estimation, in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 2850–2859.
- Benchmarking of a full-body inertial motion capture system for clinical gait analysis. 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society , 4579–4582.
- Markerless motion capture through visual hull, articulated icp and subject specific model generation. International Journal of Computer Vision 87, 156–169.
- Opensim: Open-source software to create and analyze dynamic simulations of movement. IEEE Transactions on Biomedical Engineering 54, 1940–1950.
- The Construction Chart Book: The US Construction Industry and Its Workers. CPWR.
- Markerless human motion analysis in gauss–laguerre transform domain: An application to sit-to-stand in young and elderly people. IEEE Transactions on Information Technology in Biomedicine 13, 207–216.
- Flex: Parameter-free multi-view 3d human motion reconstruction. ArXiv abs/2105.01937.
- Muscle contributions to propulsion and support during running. Journal of biomechanics 43 14, 2709–16.
- Epipolar transformers. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 7776–7785.
- Human3.6m: Large scale datasets and predictive methods for 3d human sensing in natural environments. IEEE Transactions on Pattern Analysis and Machine Intelligence 36, 1325–1339.
- Learnable triangulation of human pose. 2019 IEEE/CVF International Conference on Computer Vision (ICCV) , 7717–7726.
- End-to-end recovery of human shape and pose. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition , 7122–7131.
- Adam: A method for stochastic optimization. CoRR abs/1412.6980.
- Vibe: Video inference for human body pose and shape estimation. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 5252–5262.
- Hybrik: A hybrid analytical-neural inverse kinematics solution for 3d human pose and shape estimation. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 3382–3392.
- Smpl: a skinned multi-person linear model. ACM Trans. Graph. 34, 248:1–248:16.
- A simple yet effective baseline for 3d human pose estimation. 2017 IEEE International Conference on Computer Vision (ICCV) , 2659–2668.
- A deep neural network-based method for estimation of 3d lifting motions. Journal of biomechanics 84, 87–93.
- Using a marker-less method for estimating l5/s1 moments during symmetrical lifting. Applied ergonomics 65, 541–550.
- Development of an optical motion-capture system for 3d gait analysis. 2011 2nd International Conference on Instrumentation, Communications, Information Technology, and Biomedical Engineering , 391–394.
- V2v-posenet: Voxel-to-voxel prediction network for accurate 3d hand and human pose estimation from a single depth map. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition , 5079–5088.
- Stacked hourglass networks for human pose estimation, in: ECCV.
- Coarse-to-fine volumetric prediction for single-image 3d human pose. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , 1263–1272.
- Quaternet: A quaternion-based recurrent model for human motion. ArXiv abs/1805.06485.
- Imagenet large scale visual recognition challenge. International Journal of Computer Vision 115, 211–252.
- Deep high-resolution representation learning for human pose estimation. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 5686–5696.
- Deeppose: Human pose estimation via deep neural networks. 2014 IEEE Conference on Computer Vision and Pattern Recognition , 1653–1660.
- Deep kinematic pose regression, in: ECCV Workshops.
- On the continuity of rotation representations in neural networks. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 5738–5746.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.