Diagnosing Error in Temporal Action Detectors

TL;DR

This paper introduces a diagnostic tool for analyzing the performance of temporal action detectors in videos, providing insights beyond single metrics and highlighting key areas for improvement in the field.

Contribution

The paper presents a new diagnostic tool for temporal action detection analysis and demonstrates its use on recent challenge results, offering detailed insights into performance issues.

Findings

Handling temporal context is crucial for improvement.

Robustness to instance size variations impacts detection accuracy.

Reducing localization errors significantly enhances performance.

Abstract

Despite the recent progress in video understanding and the continuous rate of improvement in temporal action localization throughout the years, it is still unclear how far (or close?) we are to solving the problem. To this end, we introduce a new diagnostic tool to analyze the performance of temporal action detectors in videos and compare different methods beyond a single scalar metric. We exemplify the use of our tool by analyzing the performance of the top rewarded entries in the latest ActivityNet action localization challenge. Our analysis shows that the most impactful areas to work on are: strategies to better handle temporal context around the instances, improving the robustness w.r.t. the instance absolute and relative size, and strategies to reduce the localization errors. Moreover, our experimental analysis finds the lack of agreement among annotator is not a major roadblock to attain progress in the field. Our diagnostic tool is publicly available to keep fueling the minds of other researchers with additional insights about their algorithms.