从关键点到预测分布：YOLO-Pose模型的事后不确定性
From Keypoints to Predictive Distributions: Post-Hoc Uncertainty for YOLO-Pose Models
浏览论文内容
中文总结 AI 辅助
本研究针对YOLO-Pose模型未量化空间不确定性的问题，提出轻量级事后概率扩展方法，结合校准诊断与AKP的评估协议，经COCO实验验证其在关键点可靠性排序等任务中有效，并在飞机着陆场景得到应用。
中文摘要 AI 辅助
YOLO-Pose模型可实现高效的关键点定位，但未量化相关的空间不确定性。我们提出一种轻量级事后概率扩展方法，在已训练的YOLO-Pose模型基础上，为每个关键点位置增加以模型原始预测为中心的校准双变量预测分布。具体而言，我们训练额外的概率头，采用重要性加权负对数似然来预测每个关键点的输入依赖2×2分散矩阵，可选择高斯校准以适配广泛下游任务，或Student-t校准以保证分布保真度。作为补充，我们提出评估协议，将一系列分布校准诊断指标与平均关键点精度（AKP）相结合，AKP是COCO AP协议在关键点层面的扩展，用于评估可靠性排序。在COCO数据集上的实验表明，学习到的不确定性估计能实现有效的关键点层面可靠性排序，Student-t校准最能拟合经验残差分布，基于不确定性的剪枝可移除不可靠关键点。一项核心应用演示是基于视觉的飞机着陆，其中跑道关键点的校准协方差支持不确定性感知的飞机位置估计及下游传感器融合。
英文摘要
YOLO-Pose models provide efficient keypoint localization, but do not quantify the associated spatial uncertainty. We introduce a lightweight post-hoc probabilistic extension that augments a trained YOLO-Pose model with calibrated bivariate predictive distributions over keypoint locations, centered at the model's original predictions. Concretely, we train additional probabilistic heads with an importance-weighted negative log-likelihood to predict an input-dependent $2\times2$ dispersion matrix for each keypoint, followed by Gaussian calibration for broad downstream compatibility or Student-$t$ calibration for distributional fidelity. Complementing this, we propose an evaluation protocol that combines a suite of distributional calibration diagnostics with average keypoint precision (AKP), a keypoint-level extension of the COCO AP protocol for assessing reliability rankings. Experiments on COCO show that the learned uncertainty estimates enable effective keypoint-level reliability ranking, Student-$t$ calibration best captures the empirical residual distribution, and uncertainty-based pruning removes unreliable keypoints. A central application-level demonstration is vision-based aircraft landing, where calibrated covariances for runway keypoints support uncertainty-aware aircraft position estimation and downstream sensor fusion.
