Filters








42 Hits in 5.3 sec

LCR-Net++: Multi-person 2D and 3D Pose Detection in Natural Images [article]

Gregory Rogez, Philippe Weinzaepfel, Cordelia Schmid
2019 IEEE Transactions on Software Engineering   accepted
We propose an end-to-end architecture for joint 2D and 3D human pose estimation in natural images.  ...  Moreover, it shows promising results on real images for both single and multi-person subsets of the MPII 2D pose benchmark and demonstrates satisfying 3D pose results even for multi-person images.  ...  This work was supported by ERC advanced grant Allegro and an Amazon Academic Research Award. We thank NVIDIA for donating the GPUs used for this research.  ... 
doi:10.1109/tpami.2019.2892985 pmid:30640602 arXiv:1803.00455v3 fatcat:jslswqgavzcmjcwyp5lmiucsxq

LCR-Net: Localization-Classification-Regression for Human Pose

Gregory Rogez, Philippe Weinzaepfel, Cordelia Schmid
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose an end-to-end architecture for joint 2D and 3D human pose estimation in natural images.  ...  pose proposals; and 3) a regressor that refines pose proposals both in 2D and 3D.  ...  This work was supported by ERC advanced grant Allegro and an Amazon gift. We thank NVIDIA for donating the GPUs used for this research.  ... 
doi:10.1109/cvpr.2017.134 dblp:conf/cvpr/RogezWS17 fatcat:lwvkswnrbzgvnmri2nbsjpihty

Single-Shot Multi-Person 3D Pose Estimation From Monocular RGB [article]

Dushyant Mehta, Oleksandr Sotnychenko, Franziska Mueller, Weipeng Xu, Srinath Sridhar, Gerard Pons-Moll, Christian Theobalt
2018 arXiv   pre-print
To further stimulate research in multi-person 3D pose estimation, we will make our new datasets, and associated code publicly available for research purposes.  ...  We propose a new single-shot method for multi-person 3D pose estimation in general scenes from a monocular RGB camera.  ...  Sequence-wise evaluation of our method and LCR-net [49] on multi-person 3D pose test set MuPoTS-3D.  ... 
arXiv:1712.03453v3 fatcat:pnhwtrnqsbhelen47etpk4namm

Exploring Severe Occlusion: Multi-Person 3D Pose Estimation with Gated Convolution [article]

Renshu Gu, Gaoang Wang, Jenq-Neng Hwang
2020 arXiv   pre-print
In this paper, we propose a temporal regression network with a gated convolution module to transform 2D joints to 3D and recover the missing occluded joints in the meantime.  ...  3D human pose estimation (HPE) is crucial in many fields, such as human behavior analysis, augmented reality/virtual reality (AR/VR) applications, and self-driving industry.  ...  They further extend LCR-net to LCR-net++ and adds additional synthetic training data and show better performance on partially occluded or truncated persons in [8] .  ... 
arXiv:2011.00184v1 fatcat:6az3ghqpxfbczonp22nx4dqxpi

Real-Time Hybrid Mapping of Populated Indoor Scenes using a Low-Cost Monocular UAV [article]

Stuart Golodetz, Madhu Vankadari, Aluna Everitt, Sangyun Shin, Andrew Markham, Niki Trigoni
2022 arXiv   pre-print
However, despite many recent works on both marker-based and markerless multi-UAV single-person motion capture, markerless single-camera multi-person 3D human pose estimation remains a much earlier-stage  ...  In this paper, we present what is thus, to our knowledge, the first system to perform simultaneous mapping and multi-person 3D human pose estimation from a monocular camera mounted on a single UAV.  ...  We thank Graham Taylor for the use of the Wytham Flight Lab, Philip Torr for the use of an Asus ZenFone AR, and Tommaso Cavallari for implementing TangoCapture.  ... 
arXiv:2203.02453v1 fatcat:hs2ymgqzyfethl6eszh4iemhcm

3D Pose Detection of Closely Interactive Humans Using Multi-View Cameras

Xiu Li, Zhen Fan, Yebin Liu, Yipeng Li, Qionghai Dai
2019 Sensors  
We propose a method to automatically detect 3D poses of closely interactive humans from sparse multi-view images at one time instance.  ...  To solve this problem, we first obtain 2D joints in every image using OpenPose and human semantic segmentation results from Mask R-CNN.  ...  Similar to top-bottom approach in 2D pose detection from one single image, Rogez et al. [27, 28] regressed multi-person 3D poses by LCR-net.  ... 
doi:10.3390/s19122831 pmid:31242651 pmcid:PMC6631029 fatcat:msi6lhob3bfhtc74qjou4fqh6e

UniPose: Unified Human Pose Estimation in Single Images and Videos [article]

Bruno Artacho, Andreas Savakis
2020 arXiv   pre-print
single person pose detection for both single images and videos.  ...  Additionally, our method is extended to UniPose-LSTM for multi-frame processing and achieves state-of-the-art results for temporal pose estimation in Video.  ...  Finally, LCR-Net performs a regression to estimate 3D coordinates in the image.  ... 
arXiv:2001.08095v1 fatcat:m4gplonkgjbvfadpaokvpam4c4

DOPE: Distillation Of Part Experts for whole-body 3D pose estimation in the wild [article]

Philippe Weinzaepfel, Romain Brégier, Hadrien Combaluzier, Vincent Leroy, Grégory Rogez
2020 arXiv   pre-print
designed for whole-body 2D-3D pose detection.  ...  In practice, given a training image with partial or no annotation, each part expert detects its subset of keypoints in 2D and 3D and the resulting estimations are combined to obtain whole-body pseudo ground-truth  ...  For our body expert, we employ LCR-Net++ [55] that jointly estimates 2D and 3D body poses from the image and has demonstrated robustness to challenging in-the-wild scenarios, i.e., showing multiple interacting  ... 
arXiv:2008.09457v1 fatcat:t7w5vcg2bzg5vc2fkgysyyst6y

Absolute Human Pose Estimation with Depth Prediction Network [article]

Márton Véges, András Lőrincz
2019 arXiv   pre-print
Unlike previous methods, our network works in a single step without any post-processing. Our network beats previous methods on the MuPoTS-3D dataset and achieves state-of-the-art results.  ...  This works well for a single person but is insufficient in the case of multiple interacting people.  ...  The 2D pose detector is the state-of-theart multi-person pose detector OpenPose [1] . In our network, the 2D-to-3D component is called 3D PoseNet.  ... 
arXiv:1904.05947v1 fatcat:oy2taji4ujcrtfgzqpnnyreg2m

UniPose: Unified Human Pose Estimation in Single Images and Videos

Bruno Artacho, Andreas Savakis
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
single person pose detection for both single images and videos.  ...  Additionally, our method is extended to UniPose-LSTM for multi-frame processing and achieves state-of-the-art results for temporal pose estimation in Video.  ...  Acknowledgements This research was supported in part by the National Science Foundation grant ♯1749376.  ... 
doi:10.1109/cvpr42600.2020.00706 dblp:conf/cvpr/ArtachoS20 fatcat:544vyz7fyfg4viqn5llskbetzu

iMapper: Interaction-guided Joint Scene and Human Motion Mapping from Monocular Videos [article]

Aron Monszpart, Paul Guerrero, Duygu Ceylan, Ersin Yumer, Niloy J. Mitra
2018 arXiv   pre-print
and 3D pose recovery.  ...  While the problem remains a subject of active research, concurrent advances have been made in the context of human pose reconstruction from monocular video, including image-space feature point detection  ...  B CONFIDENCE OF KEYPOINT DETECTION USING LCR-NET++ When using LCR-Net++, we confidence estimate 2D detection keypoints as where var k denotes the variance of the 3D joint position among the grouped pose  ... 
arXiv:1806.07889v1 fatcat:dt25djebsvcf3ek5swwpqyfvqy

HMOR: Hierarchical Multi-Person Ordinal Relations for Monocular Multi-Person 3D Pose Estimation [article]

Jiefeng Li, Can Wang, Wentao Liu, Chen Qian, Cewu Lu
2020 arXiv   pre-print
Remarkable progress has been made in 3D human pose estimation from a monocular RGB camera. However, only a few studies explored 3D multi-person cases.  ...  The proposed method significantly outperforms state-of-the-art methods on publicly available multi-person 3D pose datasets.  ...  This work is supported in part by the National Key R&D Program of China, No. 2017YFA0700800, National Natural Science Foundation of China under Grants 61772332Shanghai Qi Zhi Institute.  ... 
arXiv:2008.00206v2 fatcat:bgdxil5v45c5tf6ufuw7f5zl7u

Multi-Person Absolute 3D Human Pose Estimation with Weak Depth Supervision [article]

Marton Veges, Andras Lorincz
2020 arXiv   pre-print
This is especially true for multi-person 3D pose estimation, where, to our knowledge, there are only machine generated annotations available for training.  ...  Our algorithm is a monocular, multi-person, absolute pose estimator. We evaluate the algorithm on several benchmarks, showing a consistent improvement in error rates.  ...  Acknowledgment MV received support from the European Union and co-financed by the European Social Fund (EFOP-3.6.3-16-2017-00002).  ... 
arXiv:2004.03989v1 fatcat:suyi32ydmjgfbdgyn3v75e7sxq

Recent Advances in Monocular 2D and 3D Human Pose Estimation: A Deep Learning Perspective [article]

Wu Liu, Qian Bao, Yu Sun, Tao Mei
2021 arXiv   pre-print
2D and 3D, and the complex multi-person scenarios.  ...  Recently, benefited from the deep learning technologies, a significant amount of research efforts have greatly advanced the monocular human pose estimation both in 2D and 3D areas.  ...  Multi-person Top-down methods • LCR-Net++ [172] ; • 3DMPPE [173] .  ... 
arXiv:2104.11536v1 fatcat:tdag2jq2vjdrjekwukm5nu7l6a

Dynamic Graph Reasoning for Multi-person 3D Pose Estimation [article]

Zhongwei Qiu, Qiansheng Yang, Jian Wang, Dongmei Fu
2022 arXiv   pre-print
Multi-person 3D pose estimation is a challenging task because of occlusion and depth ambiguity, especially in the cases of crowd scenes.  ...  In this paper, we propose GR-M3D, which models the Multi-person 3D pose estimation with dynamic Graph Reasoning. The decoding graph in GR-M3D is predicted instead of pre-defined.  ...  RELATED WORK 2.1 2D Human Pose Estimation Mainstream multi-person 2D pose estimation approaches include top-down and bottom-up methods.  ... 
arXiv:2207.11341v2 fatcat:fop7t27u6vbvvmmf7vpbdrrfcy
« Previous Showing results 1 — 15 out of 42 results