Towards Better Explanations for Object Detection

Van Binh Truong, Truong Thanh Hung Nguyen, Vo Thanh Khang Nguyen, Quoc Khanh Nguyen, Quoc Hung Cao
6/5/2023
cs.CVcs.AIcs.LG

No official code found. Best implementation: Binh24399/D-CLOSE (11 stars, CC0-1.0, last commit 2024-01-10).

Abstract

Recent advances in Artificial Intelligence (AI) technology have promoted their use in almost every field. The growing complexity of deep neural networks (DNNs) makes it increasingly difficult and important to explain the inner workings and decisions of the network. However, most current techniques for explaining DNNs focus mainly on interpreting classification tasks. This paper proposes a method to explain the decision for any object detection model called D-CLOSE. To closely track the model's behavior, we used multiple levels of segmentation on the image and a process to combine them. We performed tests on the MS-COCO dataset with the YOLOX model, which shows that our method outperforms D-RISE and can give a better quality and less noise explanation.

View on arXivView PDF

Code Implementations(1)

Binh24399/D-CLOSECommunity100%
CC0-1.0

[ACML2023] Towards Better Explanations for Object Detection

111C++, PythonNov 22, 20222 years agoCC0-1.0

Similar Papers

Explaining What Machines See: XAI Strategies in Deep Object Detection Models

88% match2025

G-CAME: Gaussian-Class Activation Mapping Explainer for Object Detectors

87% match1 repo32023

Two-Stage Holistic and Contrastive Explanation of Image Classification

86% match1 repo102023

SSD: Single Shot MultiBox Detector

86% match1 repo4.8k2015

PiPS: Post-Hoc Prototypical Explanations for Interpretable Semantic Segmentation

85% match1 repo2026

Leveraging Activations for Superpixel Explanations

85% match2024

Cite this paper

@article{truong2023towards,
  title  = {Towards Better Explanations for Object Detection},
  author = {Van Binh Truong and Truong Thanh Hung Nguyen and Vo Thanh Khang Nguyen and Quoc Khanh Nguyen and Quoc Hung Cao},
  year   = {2023},
  eprint = {2306.02744},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2306.02744v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2306.02744.svg)](https://codeofpaper.com/paper/2306.02744)

Discussion