首页|期刊导航|实验技术与管理|基于单目深度先验与特征一致性的域泛化立体匹配方法

基于单目深度先验与特征一致性的域泛化立体匹配方法OA

Domain generalization stereo matching method using monocular depth priors and feature consistency

中文摘要英文摘要

针对合成数据集上训练的立体匹配模型在迁移至真实场景时,常因域偏移导致性能显著退化的问题,文章提出了一种基于单目深度先验引导与特征一致性约束的域泛化立体匹配方法(GCNet-FCDepth),用预训练单目深度模型(Depth Anything V2)提取冻结的多尺度几何结构特征,再通过选择性通道融合模块将此特征与纹理特征融合,构建深度感知的匹配特征;采用了深度引导的动态代价聚合模块,根据视差假设与深度特征的亲和度,自适应地生成聚合权重;在训练阶段引入立体对比损失与选择性白化损失,显式地约束匹配特征的域不变性.结果表明,零样本跨域泛化实验中,该方法在 ETH3D 数据集上的错误率较当前最优实时方法 RT-IGEV 降低约 64%;所提方法在 KITTI 2015 基准测试中 D1-all 达 0.72,综合性能优于所有对比方法,测距相对误差最大值不超过 3%,可满足实际场景的部署要求.

[Objective]Stereo vision simulates human binocular perception of scene depth and plays a critical role in applications such as autonomous driving,robotic navigation,and 3D reconstruction.Its key step,stereo matching,aims to establish pixel-wise correspondences between left and right views.To date,deep learning-based stereo matching methods have achieved excellent performance on standard datasets.However,due to the high cost of acquiring ground-truth disparity in real-world scenes,existing methods are often trained on synthetic datasets(e.g.,SceneFlow).When these models are used in real-world scenarios,they often suffer from domain shift caused by factors such as differences in lighting conditions,variations in color and contrast,and scene types,leading to substantial performance degradation.To address this issue,this study investigates single-domain generalization in stereo matching.[Methods]This study proposes a domain-generalized stereo matching method based on depth prior guidance and feature consistency constraints(GCNet-FCDepth).The proposed method leverages the pretrained monocular depth model Depth Anything V2 to extract frozen multiscale geometric structural features,effectively capturing scene geometry without requiring additional stereo supervision.These geometric features are then fused with texture features via a selective channel fusion module,which adaptively selects the most informative channels from both feature streams to construct depth-aware matching features.This fusion strategy ensures that the resulting representation retains robust structural cues while remaining sensitive to surface details.To further enhance the matching process,a depth-guided dynamic cost aggregation module is designed.This module adaptively generates aggregation weights based on the affinity between disparity hypotheses and depth features.By incorporating geometric priors into the aggregation process,the module improves the reliability of the cost volume,especially in textureless regions and near depth discontinuities.During training,two complementary loss functions are introduced to explicitly constrain the domain invariance of matching features.The stereo contrastive loss encourages feature consistency between corresponding pixels across stereo views while suppressing irrelevant variations.The selective whitening loss promotes feature decorrelation and reduces domain-specific biases,thereby improving generalization across unseen environments.[Results]Experimental results demonstrate that in zero-shot cross-domain generalization experiments,the error rate on the ETH3D dataset is reduced by approximately 64%relative to that of the current state-of-the-art real-time method RT-IGEV.On the KITTI 2015 benchmark,the proposed method achieves a D1-all score of 0.72 surpassing all compared methods in overall performance.[Conclusions]The proposed stereo matching method exhibits strong robustness in adverse weather scenarios such as fog and rain,thereby meeting the deployment requirements of real-world applications.

都双丽;张奕凡;王静

西安理工大学 计算机科学与工程学院,陕西 西安 710048西安理工大学 计算机科学与工程学院,陕西 西安 710048西安理工大学 计算机科学与工程学院,陕西 西安 710048

交通工程

立体匹配域泛化深度先验特征一致性

stereo matchingdomain generalizationdepth priorfeature consistency

《实验技术与管理》 2026 (10)

58-66,9

国家自然科学基金项目(62571425)陕西省自然科学基金项目(2024JC-ZDXM-35)陕西省科协青年人才托举计划(20240146)中国高校产学研创新基金-新一代信息技术创新项目(2024IT088)

10.16791/j.cnki.sjg.2026.10.008

评论