Benchmarking and Error Diagnosis in Multi-Instance Pose Estimation

TL;DR

This paper introduces a new benchmarking method and error analysis framework for multi-instance pose estimation algorithms, revealing how various factors influence their performance and error types.

Contribution

It presents a novel error diagnosis technique and benchmark for comparing pose estimation algorithms, focusing on error types and their dependence on instance attributes.

Findings

Performance varies with instance size, type, and clutter.

Localization and background errors are most affected by clutter.

Algorithms are most sensitive to the number of visible keypoints.

Abstract

We propose a new method to analyze the impact of errors in algorithms for multi-instance pose estimation and a principled benchmark that can be used to compare them. We define and characterize three classes of errors - localization, scoring, and background - study how they are influenced by instance attributes and their impact on an algorithm's performance. Our technique is applied to compare the two leading methods for human pose estimation on the COCO Dataset, measure the sensitivity of pose estimation with respect to instance size, type and number of visible keypoints, clutter due to multiple instances, and the relative score of instances. The performance of algorithms, and the types of error they make, are highly dependent on all these variables, but mostly on the number of keypoints and the clutter. The analysis and software tools we propose offer a novel and insightful approach for understanding the behavior of pose estimation algorithms and an effective method for measuring their strengths and weaknesses.