半稠密匹配的不确定性不只是局部置信度
Semi-Dense Matching Uncertainty Is Not Just Local Confidence
浏览论文内容
中文总结 AI 辅助
针对半稠密匹配不确定性量化不足的问题，本文提出含9个可学习参数的双分量校准拉普拉斯混合模型框架与CoRe几何重拟合模块，显著提升了下游几何估计精度且计算开销极小。
中文摘要 AI 辅助
可靠的半稠密匹配是现代几何视觉系统的核心需求，这类方法通常采用由粗到精的范式，在性能与计算成本间实现最优平衡。然而现有方法往往难以提供准确量化的不确定性，会忽略灾难性的粗匹配分配失败，导致误差分布被截断，严重误判几何估计结果。本文提出一种轻量级事后整体不确定性估计框架，引入仅含9个可学习参数的双分量校准拉普拉斯混合模型，目标是明确捕捉精细局部优化的尖锐噪声与粗匹配分配失败的宽尾部。我们提出Coarse-success后验重拟合（CoRe）方法，这是一个几何重拟合模块，利用粗匹配成功的后验概率作为软对应权重。大量实验表明，该方法在各类仅预训练匹配器与鲁棒估计器的下游几何精度上均有提升，且计算开销极小，代码可在指定网址获取。
英文摘要
Reliable semi-dense matching is essential for modern geometric vision systems. Designed under a coarse-to-fine paradigm, it achieves an optimal balance between performance and computational cost. However, existing methods often struggle to provide well-quantified uncertainties, where catastrophic coarse-assignment failures are ignored, leading to truncated error distributions and severely misjudged geometric estimations. In this paper, we propose a lightweight, post-hoc overall uncertainty estimation framework that introduces a two-component calibrated Laplace mixture model with only 9 learnable parameters. The objective is to explicitly capture both the sharp local refinement noise and the broader tail of coarse-assignment failures. We introduce the Coarse-success posterior Refit (CoRe) method, a geometric refitting module that utilizes the posterior probability of coarse-assignment success as soft correspondence weights. Extensive experiments show that our method consistently improves downstream geometric accuracy across various pretrained-only matchers and robust estimators with minimal computational overhead. Our code is available at https://github.com/khoavpt/Probabilistic-matching.
发表机构
- University of Engineering and Technology, Vietnam National University（越南国家大学工程技术大学）
- Optoelectronics Center, Viettel Aerospace Institute, Viettel Group（Viettel集团Viettel宇航研究院光电中心）
- National Yang Ming Chiao Tung University（国立阳明交通大学）
机构由 AI 辅助整理，请以论文原文为准。
