Conditional regression forests for human pose estimation

By Min Sun, Pushmeet Kohli, Jamie Shotton. 2012

193 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “[54] Conditional regression forests Single depth image 3D skeleton, over 80% average precision Charles and Everingham [55] Limb-based shape models Single depth image 2D skeleton, robust to occlusions”
Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, neutral): “[38] also incorporate prior information (e.”
Flowing ConvNets for Human Pose Estimation in Videos (2015, neutral): “More recent work has tackled pose estimation holistically: initially with Random Forests on depth data [12, 29, 31, 34] and RGB [3, 4, 24], and most recently with convolutional neural networks.”
Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture (2014, methodology): “Furthermore, few bottom-up approaches use a global refinement step, such as enforcing dependency between local outputs [16] or kinematic constraints [17].”
Effective 3D action recognition using EigenJoints (2014, methodology): “With the release of RGBD cameras and the associated SDK, we are able to recover 3D positions of skeleton joints in real time and with reasonable accuracy [7,20,21].”
RMPE: Regional Multi-person Pose Estimation (2016, methodology): “on is a fundamental challenge for computer vision. In practice, recognizing the pose of multiple persons in the wild is a lot more challenging than recognizing the pose of a single person in an image [27, 28, 18, 20, 35]. Recent attempts approach this problemby using either a two-step framework [25,”
Pictorial Human Spaces: How Well Do Humans Perceive a 3D Articulated Pose? (2013, methodology): “Instead, qualitative representations may be used for most tasks, although the implications in skill games (e.g. using Kinect[23]) where player’s accuracy is valued, but may not be realizable, could be relevant.”
Random tree walk toward instantaneous 3D human pose estimation (2015, methodology): “Using the RTW, large computational gain is possible compared to the previous methods [19, 13, 20] because the regression tree is iteratively evaluated per joint, rather than per pixel.”
Real-time facial feature detection using conditional regression forests (2012, methodology): “1Independent of this work, a similar concept has been applied in the context of human pose estimation [28] 3.”
Cascaded hand pose regression (2015, methodology): “This framework has been applied to facial landmark localization [8], human body pose estimation [33] and hand pose estimation [6, 10].”
AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time (2022, background): “In conventional body-only pose estimation, recognizing the pose of multiple persons in the wild is more challenging than recognizing the pose of a single person in an image [4], [5], [6], [7], [8].”
Hand Pose Estimation through Weakly-Supervised Learning of a Rich Intermediate Representation (2015, methodology): “Experiments on the NYU dataset [32] show that the proposed training method decreases error on joints over direct regression of joints from depth data by 15.7%.”
Pose for Action - Action for Pose (2016, methodology): “Random regression forests have been proven to be robust for the task of human pose estimation [5], [24], [27].”
Real-Time Articulated Hand Pose Estimation Using Semi-supervised Transductive Regression Forests (2013, background): “While latest depth sensor technology has enabled body pose estimation in real-time [2, 24, 12, 26], hand pose estimation still requires improvement.”
A Comprehensive Analysis of Deep Regression (2018, unknown): “In computer vision, regression techniques span a large ensemble of applicative scenarios such as: head-pose estimation [1], [2], facial landmark detection [3], [4], human pose estimation [5], [6], age estimation [7], [8], or image registration [9], [10].”
Accurate, Robust, and Flexible Real-time Hand Tracking (2015, methodology): “Second, while following existing practice [25, 12] in breaking the regression into two ‘layers’ (or stages), we focus the first layer exclusively in predicting quantized global hand rotation.”
Video Object Segmentation without Temporal Information (2017, background): “Similarly, [64] proposes to estimate human pose dependent on torso orientation, or human height, which can be a useful cue for the task of pose estimation.”
Uncertainty-Driven 6D Pose Estimation of Objects and Scenes from a Single RGB Image (2016, background): “item 3, is of interest to all methods where the final optimization depends on variables (explicitly or latent) such as object instance [2], person height [25] or object part [19].”