方法对比

	可解释目标检测	可解释视觉 Transformer
领域	深度学习	深度学习
方法族	Machine learning	Machine learning
起源年份	2016–2017	2021
提出者	Selvaraju et al. (Grad-CAM); Ribeiro et al. (LIME); Lundberg & Lee (SHAP)	Chefer, H., Gur, S., & Wolf, L. (attribution framework); Dosovitskiy et al. (base ViT)
类型	Post-hoc explainability applied to object detection	Post-hoc explainability applied to Vision Transformer
开创性文献	Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization. Proceedings of the IEEE International Conference on Computer Vision (ICCV), 618–626.	Chefer, H., Gur, S., & Wolf, L. (2021). Transformer interpretability beyond attention visualization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 782–791.
别名	XAI Object Detection, Interpretable Object Detection, Transparent Object Detection, Explainable OD	XViT, Interpretable ViT, Explainable ViT, Transparent Vision Transformer
相关	5	5
摘要	Explainable object detection combines a deep-learning object detector — such as YOLO, Faster R-CNN, or DETR — with post-hoc or built-in explainability methods (Grad-CAM, LIME, SHAP, D-RISE) that visualize why the model placed a bounding box at a particular location and assigned a particular class label, making its decisions auditable by humans.	Explainable Vision Transformer combines the strong image-recognition performance of Vision Transformers (ViT) with attribution techniques — such as relevance propagation, attention rollout, or gradient-weighted attention — that highlight which image regions drive each prediction. The approach enables researchers and practitioners to audit model decisions and satisfy transparency requirements without sacrificing accuracy.