Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pan, Baiyu, jiao, jichao, Yao, Bowen, Pang, Jianxin, Cheng, Jun
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2410.06527
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914968198184960
author Pan, Baiyu
jiao, jichao
Yao, Bowen
Pang, Jianxin
Cheng, Jun
author_facet Pan, Baiyu
jiao, jichao
Yao, Bowen
Pang, Jianxin
Cheng, Jun
contents The soft-argmax operation is widely adopted in neural network-based stereo matching methods to enable differentiable regression of disparity. However, network trained with soft-argmax is prone to being multimodal due to absence of explicit constraint to the shape of the probability distribution. Previous methods leverages Laplacian distribution and cross-entropy for training but failed to effectively improve the accuracy and even compromises the efficiency of the network. In this paper, we conduct a detailed analysis of the previous distribution-based methods and propose a novel supervision method for stereo matching, Sampling-Gaussian. We sample from the Gaussian distribution for supervision. Moreover, we interpret the training as minimizing the distance in vector space and propose a combined loss of L1 loss and cosine similarity loss. Additionally, we leveraged bilinear interpolation to upsample the cost volume. Our method can be directly applied to any soft-argmax-based stereo matching method without a reduction in efficiency. We have conducted comprehensive experiments to demonstrate the superior performance of our Sampling-Gaussian. The experimental results prove that we have achieved better accuracy on five baseline methods and two datasets. Our method is easy to implement, and the code is available online.
format Preprint
id arxiv_https___arxiv_org_abs_2410_06527
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The Sampling-Gaussian for stereo matching
Pan, Baiyu
jiao, jichao
Yao, Bowen
Pang, Jianxin
Cheng, Jun
Computer Vision and Pattern Recognition
Artificial Intelligence
The soft-argmax operation is widely adopted in neural network-based stereo matching methods to enable differentiable regression of disparity. However, network trained with soft-argmax is prone to being multimodal due to absence of explicit constraint to the shape of the probability distribution. Previous methods leverages Laplacian distribution and cross-entropy for training but failed to effectively improve the accuracy and even compromises the efficiency of the network. In this paper, we conduct a detailed analysis of the previous distribution-based methods and propose a novel supervision method for stereo matching, Sampling-Gaussian. We sample from the Gaussian distribution for supervision. Moreover, we interpret the training as minimizing the distance in vector space and propose a combined loss of L1 loss and cosine similarity loss. Additionally, we leveraged bilinear interpolation to upsample the cost volume. Our method can be directly applied to any soft-argmax-based stereo matching method without a reduction in efficiency. We have conducted comprehensive experiments to demonstrate the superior performance of our Sampling-Gaussian. The experimental results prove that we have achieved better accuracy on five baseline methods and two datasets. Our method is easy to implement, and the code is available online.
title The Sampling-Gaussian for stereo matching
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2410.06527