DExT: Detector Explanation Toolkit

TL;DR

DExT is an open-source toolkit that provides holistic, gradient-based explanations for object detector decisions, improving interpretability for safety-critical applications and enabling better evaluation of detector transparency.

Contribution

The paper introduces DExT, a comprehensive toolkit for explaining both bounding box and classification decisions of object detectors using gradient-based methods.

Findings

SSD is more faithfully explained than other detectors.

SmoothGrad with Guided Backpropagation offers the most trustworthy explanations.

DExT encourages interpretability evaluation of object detectors.

Abstract

State-of-the-art object detectors are treated as black boxes due to their highly non-linear internal computations. Even with unprecedented advancements in detector performance, the inability to explain how their outputs are generated limits their use in safety-critical applications. Previous work fails to produce explanations for both bounding box and classification decisions, and generally make individual explanations for various detectors. In this paper, we propose an open-source Detector Explanation Toolkit (DExT) which implements the proposed approach to generate a holistic explanation for all detector decisions using certain gradient-based explanation methods. We suggests various multi-object visualization methods to merge the explanations of multiple objects detected in an image as well as the corresponding detections in a single image. The quantitative evaluation show that the Single Shot MultiBox Detector (SSD) is more faithfully explained compared to other detectors regardless of the explanation methods. Both quantitative and human-centric evaluations identify that SmoothGrad with Guided Backpropagation (GBP) provides more trustworthy explanations among selected methods across all detectors. We expect that DExT will motivate practitioners to evaluate object detectors from the interpretability perspective by explaining both bounding box and classification decisions.