论文阅读 [CVPR-2022] Attention Concatenation Volume for Accurate and Efficient Stereo Matching
论文阅读 [CVPR-2022] Attention Concatenation Volume for Accurate and Efficient Stereo Matching
studyai.com
搜索论文: [Attention Concatenation Volume for Accurate and Efficient Stereo Matching](http://www.studyai.com/search/whole-site/?q=Attention Concatenation Volume for Accurate and Efficient Stereo Matching)
http://www.studyai.com/search/whole-site/?q=Attention+Concatenation+Volume+for+Accurate+and+Efficient+Stereo+Matching
摘要(Abstract)
Stereo matching is a fundamental building block for many vision and robotics applications.
立体匹配是许多视觉和机器人应用的基本组成部分。
An informative and concise cost volume representation is vital for stereo matching of high accuracy and efficiency.
对于高精度和高效率的立体匹配来说，信息丰富且简洁的成本-体积表示至关重要。
In this paper, we present a novel cost volume construction method which generates attention weights from correlation clues to suppress redundant information and enhance matching-related information in the concatenation volume.
在本文中，我们提出了一种新的成本量构造方法，该方法根据相关线索生成注意权重，以抑制冗余信息并增强级联量中的匹配相关信息。
To generate reliable attention weights, we propose multi-level adaptive patch matching to improve the distinctiveness of the matching cost at different disparities even for textureless regions.
为了产生可靠的注意权重，我们提出了多级自适应面片匹配，以提高匹配成本在不同差异下的显著性，即使对于无纹理区域也是如此。
The proposed cost volume is named attention concatenation volume (ACV) which can be seamlessly embedded into most stereo matching networks, the resulting networks can use a more lightweight aggregation network and meanwhile achieve higher accuracy, e.g. using only 1/25 parameters of the aggregation network can achieve higher accuracy for GwcNet.
建议的成本量被称为注意力连接量（ACV），它可以无缝嵌入到大多数立体匹配网络中，由此产生的网络可以使用更轻量级的聚合网络，同时实现更高的精度，例如，仅使用聚合网络的1/25参数就可以实现GwcNet的更高精度。
Furthermore, we design a highly accurate network (ACVNet) based on our ACV, which achieves state-of-the-art performance on several benchmarks.
此外，我们基于我们的ACV设计了一个高精度网络（ACVNet），它在多个基准上实现了最先进的性能。
The code is available at https://github.com/gangweiX/ACVNet.