Human Pose Estimation using Deep Consensus Voting

TL;DR

This paper introduces a novel human pose estimation method that uses a deep consensus voting scheme, leveraging entire image information to improve keypoint detection and joint probability estimation from a single image.

Contribution

The approach employs dense voting with CNNs for keypoints, enabling image-dependent joint probability computation, which is a novel advancement over previous relative keypoint models.

Findings

Achieves competitive results on MPII and Leeds datasets.

Utilizes whole-image voting for improved keypoint localization.

Provides a new way to estimate joint probabilities based on consensus.

Abstract

In this paper we consider the problem of human pose estimation from a single still image. We propose a novel approach where each location in the image votes for the position of each keypoint using a convolutional neural net. The voting scheme allows us to utilize information from the whole image, rather than rely on a sparse set of keypoint locations. Using dense, multi-target votes, not only produces good keypoint predictions, but also enables us to compute image-dependent joint keypoint probabilities by looking at consensus voting. This differs from most previous methods where joint probabilities are learned from relative keypoint locations and are independent of the image. We finally combine the keypoints votes and joint probabilities in order to identify the optimal pose configuration. We show our competitive performance on the MPII Human Pose and Leeds Sports Pose datasets.