View independent human body pose estimation from a single perspective image

By Vasu Parameswaran, Ramalingam Chellappa. 2004

92 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. Michael J. Black
2. Angjoo Kanazawa
3. Jitendra Malik
4. Georgios Pavlakos
5. Cordelia Schmid
6. Andrew Zisserman
7. Andrea Vedaldi
8. Luc Van Gool

Citations in the corpus

• End-to-End Recovery of Human Shape and Pose (2017, methodology): “Most methods make some assumption about the limb-length or proportions [4, 21, 32, 34].”
• Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image (2017, background): “Early approaches [19, 34, 25, 6] took advantage of anatomical knowledge of the human skeleton or joint angle limits to recover pose from a single image.”
• Exploiting Temporal Information for 3D Human Pose Estimation (2017, methodology): “Several 3D pose methods have used this representation [30, 31, 22, 10].”
• Reconstructing 3D Human Pose from 2D Image Landmarks (2012, methodology): “Parameswaran and Chellappa [11] used a strong prior on skeletal size and employed 3D model based invariants to recover the joint angle configuration but made restrictive assumptions on the 3D configurations possible.”
• Pose-conditioned joint angle limits for 3D human pose reconstruction (2015, methodology): “Parameswaran and Chellappa [21] use 3D model-based invariants to recover the joint angle configuration.”
• A Simple Yet Effective Baseline for 3d Human Pose Estimation (2017, background): “2d to 3d angular pose There is a second branch of algorithms for inferring 3d pose from images which estimate the body configuration in terms of angles (and sometimes body shape) instead of directly estimating the 3d position of the joints [4, 7, 31, 54].”
• Keep It SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image (2016, background): “Parameswaran and Chellappa [33] assume that limb lengths are isometric across people, varying only in global scaling.”
• QuaterNet: A Quaternion-based Recurrent Model for Human Motion (2018, background): “poses from images or 2D joint positions [2, 42, 45], and motion forecasting of rigid bodies.”
• Vision-based human motion analysis: An overview (2007, methodology): “In [76], no camera model is assumed but fixed segment ratios are used.”