Visualizing Object Detection Features

TL;DR

This paper presents algorithms to visualize object detection features by inverting feature representations to natural images, enabling analysis of detector failures and insights into false alarms.

Contribution

It introduces a novel visualization method for object detection features that reveals the causes of false alarms and enhances understanding of detection systems.

Findings

False alarms resemble true positives in feature space

Feature visualization exposes limitations of current feature representations

Insights suggest improving features may reduce errors

Abstract

We introduce algorithms to visualize feature spaces used by object detectors. Our method works by inverting a visual feature back to multiple natural images. We found that these visualizations allow us to analyze object detection systems in new ways and gain new insight into the detector's failures. For example, when we visualize the features for high scoring false alarms, we discovered that, although they are clearly wrong in image space, they do look deceptively similar to true positives in feature space. This result suggests that many of these false alarms are caused by our choice of feature space, and supports that creating a better learning algorithm or building bigger datasets is unlikely to correct these errors. By visualizing feature spaces, we can gain a more intuitive understanding of recognition systems.