多尺度几何特征的融合和频域分解用于立体匹配网络的立体匹配网络
Hua Hou1, Diancheng Wang1, Jinqian Xu1
1School of Information and Electrical Engineering, Hebei University of Engineering, Handan, China.
PloS one
|January 16, 2026
概括
本研究介绍了一种新的立体匹配网络,该网络集成了多尺度的几何特征和频域分解. 该方法提高了差异估计的准确性,特别是在物体边界和封闭区域,实现了最先进的结果.
科学领域:
- 计算机视觉 计算机视觉
- 深度学习 (Deep Learning) 是一种深度学习.
- 几何计算机视觉 几何计算机视觉
背景情况:
- 基于学习的立体匹配方法依赖于成本量来准确估计差异.
- 现有的成本量往往缺乏全球几何信息,导致前景/后景混乱和细节模糊的问题.
研究的目的:
- 提出一个立体相匹配网络,通过结合多尺度几何特征和频域分解来解决传统成本量的限制.
- 提高立体声匹配的精度和效率,特别是在边缘和遮等具有挑战性的区域.
主要方法:
- 使用多尺度几何提取模块将本地相关性转换为全球几何理解.
- 适应性道注意力机制用于高效的成本聚合.
- 差异精制利用多尺度GRU和频域分解网络进行高分辨率的差异重建.
主要成果:
- 在多个基准数据集上实现了最先进的性能 (Scene mFlow,KITTI2012,KITTI2015,ETH3D,Middlebury).
- 在KITTI2015数据集上,实现了较低的错误率:1.39% (D1-bg) 和2.54% (D1-fg).
- 保持了实时推断能力.
结论:
- 拟议的网络有效地增强了成本体积的全球几何理解.
- 多个尺度的几何特征和频域分解的融合导致了优越的立体声匹配精度和细节保存.
- 该方法为现实应用中高精度,高效的立体声匹配提供了有前途的方法.
相关概念视频
01:20Deconvolution
548
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
548
01:26Linear Approximation in Frequency Domain
358
Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
358