Multi-Person Pose Estimation Evaluation Using Optimal Transportation and Improved Pose Matching

TL;DR

This paper introduces OCpose, a novel evaluation metric for multi-person pose estimation that uses optimal transportation to fairly assess all detected poses regardless of confidence scores, providing a more balanced evaluation of true and false positives.

Contribution

The paper proposes OCpose, an evaluation metric that incorporates optimal transportation and improved pose matching to fairly evaluate all detected poses regardless of confidence scores.

Findings

OCpose offers a different perspective than traditional confidence-based metrics.

It improves the fairness of pose evaluation by considering all detections equally.

OCpose enhances the reliability of matching scores between estimated and annotated poses.

Abstract

In Multi-Person Pose Estimation, many metrics place importance on ranking of pose detection confidence scores. Current metrics tend to disregard false-positive poses with low confidence, focusing primarily on a larger number of high-confidence poses. Consequently, these metrics may yield high scores even when many false-positive poses with low confidence are detected. For fair evaluation taking into account a tradeoff between true-positive and false-positive poses, this paper proposes Optimal Correction Cost for pose (OCpose), which evaluates detected poses against pose annotations as an optimal transportation. For the fair tradeoff between true-positive and false-positive poses, OCpose equally evaluates all the detected poses regardless of their confidence scores. In OCpose, on the other hand, the confidence score of each pose is utilized to improve the reliability of matching scores between the estimated pose and pose annotations. As a result, OCpose provides a different perspective assessment than other confidence ranking-based metrics.