32,971 Hits in 5.3 sec

Deep High-resolution Network with Double Attention Residual Blocks for Human Pose Estimation

Zhanqiang Huo, Han Jin, Yingxu Qiao, Fen Luo
2020 IEEE Access  
It can be proved that the dual attention mechanism is effective for human pose estimation.  ...  See Section III D for details. Figure 2 . 2 Illustrating the Channel attention module. Figure 3 . 3 The schema of the spatial attention module.  ... 
doi:10.1109/access.2020.3044885 fatcat:sbarezuk5ffrbdnih6m3asluoq

Full-Resolution Encoder-Decoder Networks with Multi-Scale Feature Fusion for Human Pose Estimation [article]

Jie Ou, Mingjian Chen, Hong Wu
2021 arXiv   pre-print
Furthermore, we propose a novel spatial-attention-based multi-scale feature collection and distribution module (SA-MFCD) to fuse and distribute multi-scale features to boost the pose estimation.  ...  To achieve more accurate 2D human pose estimation, we extend the successful encoder-decoder network, simple baseline network (SBN), in three ways.  ...  ACKNOWLEDGEMENT This work was supported in part by the National Natural Science Foundation of China (U20B2063), the Sichuan Science and Technology Program, China (2020YFS0057), and the Fundamental Research Funds for  ... 
arXiv:2106.00566v1 fatcat:wd7k2sjvgfcjrofl72k7vgi62e

SPCNet:Spatial Preserve and Content-aware Network for Human Pose Estimation [article]

Yabo Xiao, Dongdong Yu, Xiaojuan Wang, Tianqi Lv, Yiqi Fan, Lingrui Wu
2020 arXiv   pre-print
Human pose estimation is a fundamental yet challenging task in computer vision.  ...  Extensive experiments on MPII, LSP and FLIC human pose estimation benchmarks demonstrate the effectiveness of our network.  ...  Conclusion In this paper, we propose to incorporate a Dilated Hourglass Module and a Selective Information Module into an end-to-end architecture for human pose estimation.  ... 
arXiv:2004.05834v1 fatcat:3xqj4jojqnal7p7n6jill5u67u

Efficient Human Pose Estimation by Maximizing Fusion and High-Level Spatial Attention [article]

Zhiyuan Ren, Yaohai Zhou, Yizhe Chen, Ruisong Zhou, Yayu Gao
2021 arXiv   pre-print
In this paper, we propose an efficient human pose estimation network -- SFM (slender fusion model) by fusing multi-level features and adding lightweight attention blocks -- HSA (High-Level Spatial Attention  ...  HSA learns high precise spatial information by computing the attention of spatial attention map.  ...  Attention module has the advantage to capture long-distance feature which is very important for human pose estimation task.  ... 
arXiv:2107.13693v1 fatcat:k5l5pipz7bgazgycxy6w5uvxqi

3D Human Pose Estimation with Spatial and Temporal Transformers [article]

Ce Zheng, Sijie Zhu, Matias Mendieta, Taojiannan Yang, Chen Chen, Zhengming Ding
2021 arXiv   pre-print
In this work, we present PoseFormer, a purely transformer-based approach for 3D human pose estimation in videos without convolutional architectures involved.  ...  However, in the field of human pose estimation, convolutional architectures still remain dominant.  ...  3D human pose from an intermediately estimated 2D pose.  ... 
arXiv:2103.10455v3 fatcat:bfnxjc2c7vanfkmiuxjwtl4adu

DNANet: De-Normalized Attention Based Multi-Resolution Network for Human Pose Estimation [article]

Kun Zhang, Peng He, Ping Yao, Ge Chen, Chuanguang Yang, Huimin Li, Li Fu, Tianyao Zheng
2019 arXiv   pre-print
Recently, multi-resolution networks (such as Hourglass, CPN, HRNet, etc.) have achieved significant performance on the task of human pose estimation by combining features from various resolutions.  ...  In this paper, we propose a novel type of attention module, namely De-Normalized Attention (DNA) to deal with the feature attenuations of conventional attention modules.  ...  Our approach follows the second mainstream, which extends modified HRNet with three DNA modules. Human Pose Estimation Human pose estimation remains an active topic for decades.  ... 
arXiv:1909.05090v3 fatcat:lhyfltnsxnfxphs5n3rzbvhic4

Human Mesh Recovery from Monocular Images via a Skeleton-disentangled Representation [article]

Sun Yu, Ye Yun, Liu Wu, Gao Wenpeng, Fu YiLi, Mei Tao
2019 arXiv   pre-print
In spatial, we propose an effective and pluggable "disentangling the skeleton from the details" (DSD) module.  ...  In temporal, the self-attention based temporal convolution network is proposed to efficiently exploit the short and long-term temporal cues.  ...  For instance, DSD module could be easily plugged into the existing 2D/3D pose estimation network for predicting 3D human body meshes.  ... 
arXiv:1908.07172v2 fatcat:k4ncnboleza4vc6jh3y355kgum

Augmented Parallel-Pyramid Net for Attention Guided Pose-Estimation [article]

Luanxuan Hou and Jie Cao and Yuan Zhao and Haifeng Shen and Yiping Meng and Ran He and Jieping Ye
2020 arXiv   pre-print
The target of human pose estimation is to determine body part or joint locations of each person from an image. This is a challenging problems with wide applications.  ...  To address this issue, this paper proposes an augmented parallel-pyramid net with attention partial module and differentiable auto-data augmentation.  ...  Therefore, the more important task for pose-estimation is to enhance the accuracy of the keypoints rather than involve more boxes. -Attention Partial Module.  ... 
arXiv:2003.07516v1 fatcat:p5zbegmw4zfijpqpket5avl3dm

Attention-Oriented Action Recognition for Real-Time Human-Robot Interaction [article]

Ziyang Song, Ziyi Yin, Zejian Yuan, Chong Zhang, Wanchao Chi, Yonggen Ling, Shenghao Zhang
2020 arXiv   pre-print
Specifically, a Pre-Attention network is employed to roughly focus on the interactor in the scene at low resolution firstly and then perform fine-grained pose estimation at high resolution.  ...  The other compact CNN receives the extracted skeleton sequence as input for action recognition, utilizing attention-like mechanisms to capture local spatial-temporal patterns and global semantic information  ...  TABLE I EFFICIENCY I AND ACCURACY OF DIFFERENT METHODS FOR THE INTERACTOR'S POSE ESTIMATION ON THE AID DATASET.  ... 
arXiv:2007.01065v1 fatcat:eoeqpu7p55helnwgy2rsc77cnu

Multistage Polymerization Network for Multiperson Pose Estimation

Yu-Fei Bai, Hong-Bo Zhang, Qing Lei, Ji-Xiang Du, Cong-Bin Fan
2021 Journal of Sensors  
In this paper, we propose a multistage polymerization network (MPN) for multiperson pose estimation.  ...  Multiperson pose estimation is an important and complex problem in computer vision.  ...  Acknowledgments The authors would like to thank the anonymous reviewers for their valuable and insightful comments on an earlier version of this manuscript.  ... 
doi:10.1155/2021/1484218 fatcat:fyrhvzxjazb5rj24lhhevj3ode

Combining Attention with Flow for Person Image Synthesis [article]

Yurui Ren, Yubo Wu, Thomas H. Li, Shan Liu, Ge Li
2021 arXiv   pre-print
Pose-guided person image synthesis aims to synthesize person images by transforming reference images into target poses.  ...  In this paper, we observe that the commonly used spatial transformation blocks have complementary advantages.  ...  This indicates that designing efficient spatial transformation modules is crucial for this task.  ... 
arXiv:2108.01823v1 fatcat:oyyw4xwzk5hh7draypq62mqvdi

Spatial Shortcut Network for Human Pose Estimation [article]

Te Qi , Hongtao Lu
2019 arXiv   pre-print
Luckily for pose estimation, human body is geometrically structured in images, enabling modeling of spatial dependency.  ...  In this paper, we propose a spatial shortcut network for pose estimation task, where information is easier to flow spatially.  ...  Introduction Human pose estimation is a problem with strong longrange spatial dependency.  ... 
arXiv:1904.03141v1 fatcat:dxibhdjyxzditp75lls663cmdi

Improving Human Pose Estimation with Self-Attention Generative Adversarial Networks

Xiangyang Wang, Zhongzheng Cao, Rui Wang, Zhi Liu, Xiaoqiang Zhu
2019 IEEE Access  
Some recent works try to refine the pose estimator. GAN (Generative Adversarial Networks) has been proved to be efficient to improve human pose estimation.  ...  Human pose estimation in images is challenging and important for many computer vision applications.  ...  In this paper, we introduce self-attention mechanism into Self-GAN for human pose estimation.  ... 
doi:10.1109/access.2019.2936709 fatcat:hohilrrubfcmtfk5nmwuqxo3be

Dite-HRNet: Dynamic Lightweight High-Resolution Network for Human Pose Estimation [article]

Qun Li, Ziyi Zhang, Fu Xiao, Feng Zhang, Bir Bhanu
2022 arXiv   pre-print
for human pose estimation.  ...  A high-resolution network exhibits remarkable capability in extracting multi-scale features for human pose estimation, but fails to capture long-range interactions between joints and has high computational  ...  human pose estimation.  ... 
arXiv:2204.10762v3 fatcat:2lz7oxjp6zcmzbg4gzd3cejzq4

Learning Delicate Local Representations for Multi-Person Pose Estimation [article]

Yuanhao Cai, Zhicheng Wang, Zhengxiong Luo, Binyi Yin, Angang Du, Haoqian Wang, Xiangyu Zhang, Xinyu Zhou, Erjin Zhou, Jian Sun
2020 arXiv   pre-print
To tackle this problem, we propose an efficient attention mechanism - Pose Refine Machine (PRM) to make a trade-off between local and global representations in output features and further refine the keypoint  ...  RSN aggregates features with the same spatial size (Intra-level features) efficiently to obtain delicate local representations, which retain rich low-level spatial information and result in precise keypoint  ...  It is a fundamental task for human motion recognition, kinematics analysis, human-computer interaction, animation etc. For years, human pose estimation was based on handcraft features.  ... 
arXiv:2003.04030v3 fatcat:qlm6ktpxgzaerjuszgamykyxhq
« Previous Showing results 1 — 15 out of 32,971 results