English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Vision Transformers (ViTs) have achieved state-of-the-art results on variouscomputer vision tasks, including 3D object detection. However, their end-to-endimplementation also makes ViTs less explainable, which can be a challenge fordeploying them in safety-critical applications, such as autonomous driving,where it is important for authorities, developers, and users to understand themodel's reasoning behind its predictions. In this paper, we propose a novelmethod for generating saliency maps for a DetR-like ViT with multiple camerainputs used for 3D object detection. Our method is based on the raw attentionand is more efficient than gradient-based methods. We evaluate the proposedmethod on the nuScenes dataset using extensive perturbation tests and show thatit outperforms other explainability methods in terms of visual quality andquantitative metrics. We also demonstrate the importance of aggregatingattention across different layers of the transformer. Our work contributes tothe development of explainable AI for ViTs, which can help increase trust in AIapplications by establishing more transparency regarding the inner workings ofAI models.

Explainable Multi-Camera 3D Object Detection with Transformer-Based Saliency Maps