Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations

Abstract

We present a method for estimating articulated human pose from a single static image based on a graphical model with novel pairwise relations that make adaptive use of local image measurements. More precisely, we specify a graphical model for human pose which exploits the fact the local image measurements can be used both to detect parts (or joints) and also to predict the spatial relationships between them (Image Dependent Pairwise Relations). These spatial relationships are represented by a mixture model. We use Deep Convolutional Neural Networks (DCNNs) to learn conditional probabilities for the presence of parts and their spatial relationships within image patches. Hence our model combines the representational flexibility of graphical models with the efficiency and statistical power of DCNNs. Our method significantly outperforms the state of the art methods on the LSP and FLIC datasets and also performs very well on the Buffy dataset without any training.

NIPS 2014 Camera Ready

References in corpus (2)

• DeepPose: Human Pose Estimation via Deep Neural Networks
• Detect What You Can: Detecting and Representing Objects using Holistic Models and Body Parts

Cited by in corpus (39)

• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• A Survey of Deep Learning-based Object Detection
• OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
• Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image
• Human pose estimation via Convolutional Part Heatmap Regression
• Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods
• LCR-Net++: Multi-person 2D and 3D Pose Detection in Natural Images
• AI Challenger : A Large-scale Dataset for Going Deeper in Image Understanding
• Vision-Based Assessment of Parkinsonism and Levodopa-Induced Dyskinesia with Deep Learning Pose Estimation
• Deep Learning-Based Human Pose Estimation: A Survey
• Synthesizing Training Images for Boosting Human 3D Pose Estimation
• Deep High-Resolution Representation Learning for Human Pose Estimation
• Distribution-Aware Coordinate Representation for Human Pose Estimation
• Adaptive Graphical Model Network for 2D Handpose Estimation
• Person-in-WiFi: Fine-grained Person Perception using WiFi
• Learning Temporal Pose Estimation from Sparsely-Labeled Videos
• Differentiable Hierarchical Graph Grouping for Multi-Person Pose Estimation
• Graph-PCNN: Two Stage Human Pose Estimation with Graph Pose Refinement
• Deep Metric Learning Beyond Binary Supervision
• Automatic Calibration of a Six-Degrees-of-Freedom Pose Estimation System
• Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation
• Attribute And-Or Grammar for Joint Parsing of Human Attributes, Part and Pose
• Feature Boosting Network For 3D Pose Estimation
• DeepFuse: An IMU-Aware Network for Real-Time 3D Human Pose Estimation from Multi-View Image
• From Kinematics To Dynamics: Estimating Center of Pressure and Base of Support from Video Frames of Human Motion
• Rotation-invariant Mixed Graphical Model Network for 2D Hand Pose Estimation
• Learning Dynamics from Kinematics: Estimating 2D Foot Pressure Maps from Video Frames
• Single Person Pose Estimation: A Survey
• Patch-based 3D Human Pose Refinement
• Context Modeling in 3D Human Pose Estimation: A Unified Perspective
• Fast Human Pose Estimation
• Linearized Multi-Sampling for Differentiable Image Transformation
• Key Frame Proposal Network for Efficient Pose Estimation in Videos
• Nonparametric Structure Regularization Machine for 2D Hand Pose Estimation
• Structured Context Enhancement Network for Mouse Pose Estimation
• Symmetric Non-Rigid Structure from Motion for Category-Specific Object Structure Estimation
• Semi- and Weakly-supervised Human Pose Estimation
• Layout-Graph Reasoning for Fashion Landmark Detection
• 3D Pose Estimation and Future Motion Prediction from 2D Images