Efficient and Concise Explanations for Object Detection with Gaussian-Class Activation Mapping Explainer

TL;DR

This paper introduces G-CAME, a fast and effective method for generating concise, plausible explanations for object detection models by emphasizing critical regions with Gaussian kernels, significantly reducing explanation time.

Contribution

G-CAME is a novel explanation method that produces quick, concise saliency maps for object detection, improving explanation speed and quality over existing approaches.

Findings

Reduces explanation time to 0.5 seconds

Maintains high plausibility and faithfulness of explanations

Reduces bias in tiny object detection

Abstract

To address the challenges of providing quick and plausible explanations in Explainable AI (XAI) for object detection models, we introduce the Gaussian Class Activation Mapping Explainer (G-CAME). Our method efficiently generates concise saliency maps by utilizing activation maps from selected layers and applying a Gaussian kernel to emphasize critical image regions for the predicted object. Compared with other Region-based approaches, G-CAME significantly reduces explanation time to 0.5 seconds without compromising the quality. Our evaluation of G-CAME, using Faster-RCNN and YOLOX on the MS-COCO 2017 dataset, demonstrates its ability to offer highly plausible and faithful explanations, especially in reducing the bias on tiny object detection.