通过像素集体贡献解释目标检测器
Explaining Object Detectors via Collective Contribution of Pixels
用Shapley值与交互项显式建模多个像素的集体贡献，为检测器的边界框定位与类别判定同时生成归因图，可直接迁移到其他密集预测模型（分割、关键点）的解释与调试。
原文依据 we propose a game-theoretic method based on Shapley values and interactions to explicitly capture both individual and collective pixel contributions
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论目标检测器的视觉解释对于提升其可靠性至关重要。目标检测器通过综合评估多种视觉特征来识别并定位实例。在生成解释时，若忽视检测中的这些集体影响，可能导致遗漏组合性线索或捕捉到虚假相关。然而，现有方法通常只关注单个像素的贡献，忽略了多个像素的集体贡献。为弥补这一局限，我们提出一种基于Shapley值与交互的博弈论方法，显式刻画单个像素与集体像素的贡献。该方法同时为边界框定位与类别判定提供解释，突出对检测至关重要的区域。大量实验表明，所提方法比最优方法更准确地识别出重要区域。代码见 https://github.com/tttt-0814/VX-CODE。
英文原文摘要 1001 字符
Visual explanations for object detectors are crucial for enhancing their reliability. Object detectors identify and localize instances by assessing multiple visual features collectively. When generating explanations, overlooking these collective influences in detections may lead to missing compositional cues or capturing spurious correlations. However, existing methods typically focus solely on individual pixel contributions, neglecting the collective contribution of multiple pixels. To address this limitation, we propose a game-theoretic method based on Shapley values and interactions to explicitly capture both individual and collective pixel contributions. Our method provides explanations for both bounding box localization and class determination, highlighting regions crucial for detection. Extensive experiments demonstrate that the proposed method identifies important regions more accurately than state-of-the-art methods. The code is available at https://github.com/tttt-0814/VX-CODE.