Monocular 3D Human Pose Estimation by Predicting Depth on Joints

By Xiaohan Nie, Ping Wei, Song-Chun Zhu. 2017

135 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Jun Liu
2. Wenguan Wang
3. Wanli Ouyang
4. Xiaogang Wang
5. Luc Van Gool
6. Cordelia Schmid
7. G. Wang
8. Yi Yang

Citations in the corpus

• Feature Boosting Network For 3D Pose Estimation (2019, methodology): “Previous work on 3D pose estimation has shown the benefits of using the connection information of the body parts to refine the pose estimation results or lift 2D pose to 3D space [7].”
• Hand Pose Estimation via Latent 2.5D Heatmap Regression (2018, background): “The approach [57] tries to directly regress the absolute depth from the cropped and scaled image regions which is a very ambiguous task.”
• Monocular Total Capture: Posing Face, Body, and Hands in the Wild (2018, methodology): “The two-stage methods take 2D keypoint estimation as input and focus on lifting 2D human poses to 3D without considering input image [9, 17, 33, 36, 39, 20].”
• Deep Learning-based Human Pose Estimation: A Survey (2020, methodology): “Many methods leverage prior knowledge based on the kinematic model such as skeletal joint connectivity information, joint rotation properties, and fixed bone-length ratios for plausible pose estimation, e.g., References [58, 108, 160, 173, 174, 245, 263, 309].”
• Integral Human Pose Regression (2017, methodology): “Regression methods are widely used for 3D pose estimation [38, 45, 46, 25, 27, 24, 30, 39].”
• Ordinal Depth Supervision for 3D Human Pose Estimation (2018, unknown): “[60] (ICCV’17) 90.”
• LCR-Net++: Multi-Person 2D and 3D Pose Detection in Natural Images (2018, background): “[32] predict the depth of human joints based on their 2D locations using”
• 3D Human Pose Estimation in the Wild by Adversarial Learning (2018, background): “Two-stage approaches first estimate 2D poses and then lift 2D poses to 3D poses [58, 4, 2, 51, 28, 40, 25, 57, 30].”
• Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation (2017, methodology): “Rogez and Schmid 2016; Bogo et al. 2016; Pavlakos et al. 2017; Nie, Wei, and Zhu 2017; Zhou et al. 2017; Martinez et al. 2017) and report quantitative comparisons in Table 1.”
• Monocular human pose estimation: A survey of deep learning-based methods (2020, methodology): “(Nie et al., 2017) employed LSTM to exploit global 2D joint locations and local body part images following kinematic tree body model which are two cues for joint depth estimation.”
• Single-Shot Multi-person 3D Pose Estimation from Monocular RGB (2017, background): “[38] (79.5mm) and Tome et al .”
• Deep 3D human pose estimation: A review (2021, unknown): “Early, Nie et al. (2017) process 2D poses and body part images using LSTMs, which exploit the tree structure of the human skeleton to propagate contextual information similar to Tai et al. (2015).”
• Exploiting Temporal Information for 3D Human Pose Estimation (2017, background): “With the success of deep networks, recent methods tend to focus on training a deep convolutional neural networks (CNN) end-to-end to estimate 3D poses from images directly [6,7,8,9,10,11,12,13,14,15,16].”