Structured Prediction of 3D Human Pose with Deep Neural Networks

Abstract

Most recent approaches to monocular 3D pose estimation rely on Deep Learning. They either train a Convolutional Neural Network to directly regress from image to 3D pose, which ignores the dependencies between human joints, or model these dependencies via a max-margin structured learning framework, which involves a high computational cost at inference time. In this paper, we introduce a Deep Learning regression architecture for structured prediction of 3D human pose from monocular images that relies on an overcomplete auto-encoder to learn a high-dimensional latent pose representation and account for joint dependencies. We demonstrate that our approach outperforms state-of-the-art ones both in terms of structure preservation and prediction accuracy.

References in corpus (5)

• Auto-Encoding Variational Bayes
• DeepPose: Human Pose Estimation via Deep Neural Networks
• Hands Deep in Deep Learning for Hand Pose Estimation
• Learning Human Pose Estimation Features with Convolutional Networks
• Flowing ConvNets for Human Pose Estimation in Videos

Cited by in corpus (30)

• VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera
• Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods
• A simple yet effective baseline for 3d human pose estimation
• Towards 3D Human Pose Estimation in the Wild: a Weakly-supervised Approach
• MonoPerfCap: Human Performance Capture from Monocular Video
• 3D Human Pose Estimation in the Wild by Adversarial Learning
• Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation
• 3D Human Pose Estimation with Relational Networks
• Recovering Loss to Followup Information Using Denoising Autoencoders
• Learning to Estimate 3D Human Pose and Shape from a Single Color Image
• Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimation
• Learning Human Motion Models for Long-term Predictions
• Learning to Fuse 2D and 3D Image Cues for Monocular Body Pose Estimation
• Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision
• Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose
• 3D Human Pose Estimation from a Single Image via Distance Matrix Regression
• FBI-Pose: Towards Bridging the Gap between 2D Images and 3D Human Poses using Forward-or-Backward Information
• DeepSkeleton: Skeleton Map for 3D Human Pose Regression
• PISEP^2: Pseudo Image Sequence Evolution based 3D Pose Prediction
• Straight to Shapes: Real-time Detection of Encoded Shapes
• Deep Kinematic Pose Regression
• Rethinking Pose in 3D: Multi-stage Refinement and Recovery for Markerless Motion Capture
• Task-Generic Hierarchical Human Motion Prior using VAEs
• Ordinal Depth Supervision for 3D Human Pose Estimation
• Deep Autoencoder for Combined Human Pose Estimation and body Model Upscaling
• Single-Shot Multi-Person 3D Pose Estimation From Monocular RGB
• 3D Human Pose Estimation Using Convolutional Neural Networks with 2D Pose Information
• Joint Representation of Multiple Geometric Priors via a Shape Decomposition Model for Single Monocular 3D Pose Estimation
• Explicit Spatiotemporal Joint Relation Learning for Tracking Human Pose
• 3D Human Body Reconstruction from a Single Image via Volumetric Regression