Explaining YOLO: Leveraging Grad-CAM to Explain Object Detections

11/22/2022

∙

We investigate the problem of explainability for visual object detectors. Specifically, we demonstrate on the example of the YOLO object detector how to integrate Grad-CAM into the model architecture and analyze the results. We show how to compute attribution-based explanations for individual detections and find that the normalization of the results has a great impact on their interpretation.

READ FULL TEXT

Explaining YOLO: Leveraging Grad-CAM to Explain Object Detections

The Sierpinski Object in the Scott Realizability Topos

Learning Image Aesthetic Assessment from Object-level Visual Components

Align-Deform-Subtract: An Interventional Framework for Explaining Object Differences

Black-box Explanation of Object Detectors via Saliency Maps

Improving Explainability of Disentangled Representations using Multipath-Attribution Mappings

On Explaining Your Explanations of BERT: An Empirical Study with Sequence Classification

Efficient Explanations from Empirical Explainers

Explaining YOLO: Leveraging Grad-CAM to Explain Object Detections

Related Research

The Sierpinski Object in the Scott Realizability Topos

Learning Image Aesthetic Assessment from Object-level Visual Components

Align-Deform-Subtract: An Interventional Framework for Explaining Object Differences

Black-box Explanation of Object Detectors via Saliency Maps

Improving Explainability of Disentangled Representations using Multipath-Attribution Mappings

On Explaining Your Explanations of BERT: An Empirical Study with Sequence Classification

Efficient Explanations from Empirical Explainers