Abstract:Until recently Intelligence, Surveillance, and Reconnaissance (ISR) focused on acquiring behavioral information of the targets and their activities. Continuous evolution of intelligence being gathered of the human centric activities has put increased focus on the humans, especially inferring their innate characteristics - size, shapes and physiology. These bio-signatures extracted from the surveillance sensors can be used to deduce age, ethnicity, gender and actions, and further characterize human actions in unseen scenarios. However, recovery of pose and shape of humans in such monocular videos is inherently an ill-posed problem, marked by frequent depth and view based ambiguities due to self-occlusion, foreshortening and misalignment. The likelihood function often yields a highly multimodal posterior that is difficult to propagate even using the most advanced particle filtering(PF) algorithms. Motivated by the recent success of the discriminative approaches to efficiently predict 3D poses directly from the 2D images, we present several principled approaches to integrate predictive cues using learned regression models to sustain multimodality of the posterior during tracking. Additionally, these learned priors can be actively adapted to the test data using a likelihood based feedback mechanism. Estimated 3D poses are then used to fit 3D human shape model to each frame independently for inferring anthropometric bio-signatures. The proposed system is fully automated, robust to noisy test data and has ability to swiftly recover from tracking failures even after confronting with significant errors. We evaluate the system on a large number of monocular human motion sequences.

Contour-based Iterative Pose Estimation of 3D Rigid Object

Iterative pose estimation of a 3D rigid object based on general 2D-3D contour point correspondence

Pose estimation of 3D rigid object based on tentative point correspondence

Iterative three-dimensional rigid object pose estimation with contour correspondence

Developing Rigid Constraint for the Estimation of Pose and Structure from a Single Image

ContourPose: Monocular 6D Pose Estimation Method for Reflective Texture-Less Metal Parts

Learning Stereopsis from Geometric Synthesis for 6D Object Pose Estimation

Motion Imitation of a Humanoid Robot Via Pose Estimation

A Geometric Approach to Joint 2D Region-Based Segmentation and 3D Pose Estimation Using a 3D Shape Prior

Face Pose Estimation Based on 3D Models and Affine Correspondences

An Efficient Iterative Pose Estimation Algorithm

ContourPose: Monocular 6-D Pose Estimation Method for Reflective Textureless Metal Parts.

Robust Estimation of 3D Human Poses from a Single Image

Robust Monocular 3D Car Shape Estimation from 2D Landmarks.

Robust 3D Human Pose Estimation from Single Images or Video Sequences

3D Explanation of Photo and Object Recognition Using the Constraint Satisfaction Method

Efficient and Outlier-Robust Simultaneous Pose and Correspondence Determination by Branch-and-Bound and Transformation Decomposition

Monocular 3D human pose estimation via sequential second order cone programming

Affine Correspondence Based Head Pose Estimation For A Sequence Of Images By Using A 3d Model

Object Pose Estimation Using Edge Images Synthesized from Shape Information

Coupling Top-down and Bottom-up Methods for 3D Human Pose and Shape Estimation from Monocular Image Sequences