Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking (2019-04-02T00:00:00.000000Z)

TL;DR

A Deep Conditional Variational Autoencoder based model that synthesizes diverse anatomically plausible 3D-pose samples conditioned on the estimated 2D- pose is proposed, and it is shown that CVAE-based 3d-pose sample set is consistent with the 2D to 3D lifting and helps tackling the inherent ambiguity in2D-to-3D lifting.

Abstract

Monocular 3D human-pose estimation from static images is a challenging problem, due to the curse of dimensionality and the ill-posed nature of lifting 2D-to-3D. In this paper, we propose a Deep Conditional Variational Autoencoder based model that synthesizes diverse anatomically plausible 3D-pose samples conditioned on the estimated 2D-pose. We show that CVAE-based 3D-pose sample set is consistent with the 2D-pose and helps tackling the inherent ambiguity in 2D-to-3D lifting. We propose two strategies for obtaining the final 3D pose- (a) depth-ordering/ordinal relations to score and weight-average the candidate 3D-poses, referred to as OrdinalScore, and (b) with supervision from an Oracle. We report close to state-of-the-art results on two benchmark datasets using OrdinalScore, and state-of-the-art results using the Oracle. We also show that our pipeline yields competitive results without paired image-to-3D annotations. The training and evaluation code is available at https://github.com/ssfootball04/generative_pose.

Authors

Arjun Jain

2 papers

Abhishek Sharma

2 papers

Saurabh Sharma

1 papers

TL;DR

Abstract

Authors

References61 items

Neural Body Fitting: Unifying Deep Learning and Model Based Human Pose and Shape Estimation

DRPose3D: Depth Ranking in 3D Human Pose Estimation

It's all Relative: Monocular 3D Human Pose Estimation from Weakly Supervised Data

Ordinal Depth Supervision for 3D Human Pose Estimation

A Variational U-Net for Conditional Appearance and Shape Generation

Cross-Modal Deep Variational Hand Pose Estimation

End-to-End Recovery of Human Shape and Pose

DeepSkeleton: Skeleton Map for 3D Human Pose Regression

Learning 3D Human Pose from Structure and Motion

Exploiting Temporal Information for 3D Human Pose Estimation

Integral Human Pose Regression

Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation

A Generative Model of People in Clothing

A Simple Yet Effective Baseline for 3d Human Pose Estimation

DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents

Towards 3D Human Pose Estimation in the Wild: A Weakly-Supervised Approach

Compositional Human Pose Regression

Crossing Nets: Combining GANs and VAEs with a Shared Latent Space for Hand Pose Estimation

Generating Multiple Diverse Hypotheses for Human 3D Pose Consistent with 2D Joint Detections

3D Human Pose Estimation = 2D Pose Estimation + Matching

Monocular 3D Human Pose Estimation in the Wild Using Improved CNN Supervision

3D Human Pose Estimation from a Single Image via Distance Matrix Regression

Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose

Keep It SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image

An Uncertain Future: Forecasting from Static Images Using Variational Autoencoders

DISCO Nets : DISsimilarity COefficients Networks

Single-Image Depth Perception in the Wild

Stacked Hourglass Networks for Human Pose Estimation

Convolutional Pose Machines

Deep Residual Learning for Image Recognition

Learning Structured Output Representation using Deep Conditional Generative Models

Learning Ordinal Relationships for Mid-Level Vision

Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Video

Learning Data-Driven Reflectance Priors for Intrinsic Image Decomposition

A Dual-Source Approach for 3D Pose Estimation from a Single Image

Sparse Representation for 3D Shape Estimation: A Convex Relaxation Approach

Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimation

Learning lightness from human judgement on relative reflectance

Pose-conditioned joint angle limits for 3D human pose reconstruction

DRAW: A Recurrent Neural Network For Image Generation

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Adam: A Method for Stochastic Optimization

Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments

Posebits for Monocular Human Pose Estimation

Semi-supervised Learning with Deep Generative Models

Robust Estimation of 3D Human Poses from a Single Image

2D Human Pose Estimation: New Benchmark and State of the Art Analysis

Auto-Encoding Variational Bayes

A Joint Model for 2D and 3D Pose Estimation from a Single Image

Reconstructing 3D Human Pose from 2D Image Landmarks

Deep Sparse Rectifier Neural Networks

HumanEva: Synchronized Video and Motion Capture Dataset and Baseline Algorithm for Evaluation of Articulated Human Motion

Proposal maps driven MCMC for estimating human body pose in static images

Kinematic jump processes for monocular 3D human tracking

Reconstruction of articulated objects from point correspondences in a single uncalibrated image

Smpl: A skinned multiperson linear model

Depth Sweep Regression Forests for Estimating 3D Human Pose from Images

Dropout: a simple way to prevent neural networks from overfitting

Carnegie mellon university graphics lab -motion capture library

approach –

A Global Geometric Framework for Nonlinear Dimensionality Reduction

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names