CRF-CNN: Modeling Structured Information in Human Pose Estimation

TL;DR

This paper introduces CRF-CNN, a novel neural network framework that models structural relationships in human pose estimation by integrating probabilistic message passing within CNN layers, improving accuracy on benchmark datasets.

Contribution

It presents a new CRF-CNN architecture that enables message passing between neurons in the same layer, effectively modeling structural information for human pose estimation.

Findings

Improved pose estimation accuracy on benchmark datasets.

Efficient message passing implemented with convolution operations.

End-to-end trainable CRF-CNN framework demonstrated effectiveness.

Abstract

Deep convolutional neural networks (CNN) have achieved great success. On the other hand, modeling structural information has been proved critical in many vision problems. It is of great interest to integrate them effectively. In a classical neural network, there is no message passing between neurons in the same layer. In this paper, we propose a CRF-CNN framework which can simultaneously model structural information in both output and hidden feature layers in a probabilistic way, and it is applied to human pose estimation. A message passing scheme is proposed, so that in various layers each body joint receives messages from all the others in an efficient way. Such message passing can be implemented with convolution between features maps in the same layer, and it is also integrated with feedforward propagation in neural networks. Finally, a neural network implementation of end-to-end learning CRF-CNN is provided. Its effectiveness is demonstrated through experiments on two benchmark datasets.