Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields
- Zhe Cao, Tomas Simon, Shih-En Wei, Yaser Sheikh
- Computer ScienceComputer Vision and Pattern Recognition
- 24 November 2016
We present an approach to efficiently detect the 2D pose of multiple people in an image. The approach uses a nonparametric representation, which we refer to as Part Affinity Fields (PAFs), to learn…
OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields
- Zhe Cao, Gines Hidalgo, Tomas Simon, Shih-En Wei, Yaser Sheikh
- Computer ScienceIEEE Transactions on Pattern Analysis and Machine…
- 18 December 2018
OpenPose is released, the first open-source realtime system for multi-person 2D pose detection, including body, foot, hand, and facial keypoints, and the first combined body and foot keypoint detector, based on an internal annotated foot dataset.
Convolutional Pose Machines
- Shih-En Wei, V. Ramakrishna, T. Kanade, Yaser Sheikh
- Computer ScienceComputer Vision and Pattern Recognition
- 30 January 2016
This work designs a sequential architecture composed of convolutional networks that directly operate on belief maps from previous stages, producing increasingly refined estimates for part locations, without the need for explicit graphical model-style inference in structured prediction tasks such as articulated pose estimation.
Hand Keypoint Detection in Single Images Using Multiview Bootstrapping
- Tomas Simon, H. Joo, I. Matthews, Yaser Sheikh
- Computer ScienceComputer Vision and Pattern Recognition
- 25 April 2017
An approach that uses a multi-camera system to train fine-grained detectors for keypoints that are prone to occlusion, such as the joints of a hand, and derives a result analytically relating the minimum number of views to achieve target true and false positive rates for a given detector.
Bayesian modeling of dynamic scenes for object detection
- Yaser Sheikh, M. Shah
- Computer ScienceIEEE Transactions on Pattern Analysis and Machine…
- 1 November 2005
An object detection scheme that has three innovations over existing approaches that is based on a model of the background as a single probability density, and the posterior function is maximized efficiently by finding the minimum cut of a capacitated graph.
Panoptic Studio: A Massively Multiview System for Social Interaction Capture
- H. Joo, Tomas Simon, Yaser Sheikh
- Computer ScienceIEEE Transactions on Pattern Analysis and Machine…
- 9 December 2016
The Panoptic Studio system and method are the first in reconstructing full body motion of more than five people engaged in social interactions without using markers, and empirically demonstrate the impact of the number of views in achieving this goal.
Reconstructing 3D Human Pose from 2D Image Landmarks
- V. Ramakrishna, T. Kanade, Yaser Sheikh
- BiologyEuropean Conference on Computer Vision
- 7 October 2012
This work presents an activity-independent method to recover the 3D configuration of a human figure from 2D locations of anatomical landmarks in a single image, leveraging a large motion capture corpus as a proxy for visual memory.
Nonrigid Structure from Motion in Trajectory Space
- Ijaz Akhter, Yaser Sheikh, Sohaib Khan, T. Kanade
- EngineeringNIPS
- 8 December 2008
It is shown that generic bases over trajectories, such as the Discrete Cosine Transform (DCT) basis, can be used to compactly describe most real motions.
Panoptic Studio: A Massively Multiview System for Social Motion Capture
- H. Joo, Hao Liu, Yaser Sheikh
- Computer ScienceIEEE International Conference on Computer Vision
- 7 December 2015
The Panoptic Studio is a system organized around the thesis that social interactions should be measured through the perceptual integration of a large variety of view points, consisting of integrated structural, hardware, and software innovations.
Background Subtraction for Freely Moving Cameras
- Yaser Sheikh, O. Javed, T. Kanade
- Computer ScienceIEEE International Conference on Computer Vision
- 1 September 2009
This paper extends the concept of ‘subtracting’ areas at rest to apply to video captured from a freely moving camera, and operates entirely using 2D image measurements without requiring an explicit 3D reconstruction of the scene.
...
...