Results 181 to 190 of about 18,146 (218)

On the Robustness of 3D Human Pose Estimation

2020 25th International Conference on Pattern Recognition (ICPR), 2021
It is widely shown that Convolutional Neural Networks (CNNs) are vulnerable to adversarial examples on most recognition tasks, such as image classification and segmentation. However, few work studies the more complicated task - 3D human pose estimation.
Zerui Chen, Yan Huang 0008, Liang Wang 0001   +2 more
openaire   +1 more source

3D Human Pose Estimation with 2D Human Pose and Depthmap

2020
Three-dimensional human pose estimation models are conventionally based on RGB images or by assuming that accurately-estimated (near to ground truth) 2D human pose landmarks are available. Naturally, such data only contains information about two dimensions, while the 3D poses require the three dimensions of height, width, and depth.
Zhiheng Zhou, Yue Cao, Xuanying Zhu, Henry Gardner, Hongdong Li   +4 more
openaire   +1 more source

3D human pose estimation by depth map

The Visual Computer, 2019
We present a new approach for 3D human pose estimation from a single image. State-of-the-art methods for 3D pose estimation have focused on predicting a full-body pose of a single person and have not given enough attention to the challenges in application: incompleteness of body pose and existence of multiple persons in image.
Jianzhai Wu, Dewen Hu, Fengtao Xiang, Xingsheng Yuan, Jiongming Su   +4 more
openaire   +1 more source

Learning to Augment Poses for 3D Human Pose Estimation in Images and Videos

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023
Existing 3D human pose estimation methods often suffer inferior generalization performance to new datasets, largely due to the limited diversity of 2D-3D pose pairs in the training data. To address this problem, we present PoseAug, a novel auto-augmentation framework that learns to augment the available training poses towards greater diversity and thus
Jianfeng Zhang, Kehong Gong, Xinchao Wang, Jiashi Feng   +3 more
openaire   +2 more sources

Grid Convolution for 3D Human Pose Estimation

IEEE Transactions on Pattern Analysis and Machine Intelligence
3D human pose estimation from 2D keypoint observation has been used in many human-centered computer vision applications. In this work, we tackle the task by formulating a novel grid representation learning paradigm that relies on grid convolution (GridConv), mimicking the wisdom of regular convolution operations in image space.
Yangyuxuan Kang, Dongqi Cai, Yuyang Liu, Anbang Yao, Shandong Wang, Yurong Chen 0001, Enhua Wu   +6 more
openaire   +2 more sources

Coarse-to-Fine 3D Human Pose Estimation

2019
Leveraging powerful deep convolutional networks, 2d human pose estimation has achieved great success. On the other hand, 3d human pose estimation is still a challenging task that attracts great attention. Due to the inherent depth ambiguity in 2d to 3d mapping, conventional methods are typically not able to predict 3d locations precisely, especially ...
Yu Guo 0006, Lin Zhao 0003, Shanshan Zhang 0001, Jian Yang 0003   +3 more
openaire   +1 more source

3D Human Pose Estimation with Grouping Regression

2019
Most of the methods for predicting the 3D human pose from single picture are to first extract the 2d joint position in the image, and then use the 2d joint coordinates to get the 3d joint position. This type of method focuses on learning the mapping from 2d to 3d, and neglects the kinematic properties of the joints.
Xuesheng He, Huabin Wang, Yuan Qin, Liang Tao   +3 more
openaire   +1 more source

3D Pictorial Structures for Human Pose Estimation with Supervoxels

2015 IEEE Winter Conference on Applications of Computer Vision, 2015
Pictorial structures provide a powerful framework for human pose estimation, in particular in the domain of 2D data. However, solving pictorial structures directly in 3D drastically increases its complexity and it quickly exceeds tractable dimensions.
Schick, A., Stiefelhagen, R.
openaire   +2 more sources

3D Human Body Pose Estimation by Superquadrics.

2012
This paper presents a method for 3D Human Body pose estimation. 3D real data of the searched object is acquired by a multi-camera system and segmented by a special preprocessing algorithm based on clothing analysis. The human body model is built by nine SuperQuadrics (SQ) with a-priori known anthropometric scaling and shape parameters.
I. Afanasyev, Lunardelli, Massimo, N. Biasi, Baglivo, Luca, Tavernini, Mattia, Setti, Francesco, De Cecco, Mariolino   +6 more
openaire   +2 more sources

Recurrent Transformer for 3D Human Pose Estimation

2023 4th International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE), 2023
Guang Cheng, Yan Huang, Bing Yu
openaire   +1 more source