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

TL;DR

This paper introduces a novel graphical model for articulated human pose estimation from single images, leveraging deep learning to adaptively model local image cues and spatial relationships, resulting in state-of-the-art performance.

Contribution

It combines graphical models with deep CNNs to learn image-dependent spatial relations for pose estimation, improving accuracy over previous methods.

Findings

Outperforms state-of-the-art on LSP and FLIC datasets

Performs well on Buffy dataset without training

Effectively models local image cues and spatial 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.