In recent years, convolutional neural network (CNN) algorithms promote the development of stereo matching and make great progress, but some mismatches still occur in textureless, occluded and reflective regions. In feature extraction and cost aggregation, CNNs will greatly improve the accuracy of stereo matching by utilizing global context information and high-quality feature representations. In this paper, we design a novel end-to-end stereo matching algorithm named Multi-Attention Network (MAN). To obtain the global context information in detail at the pixel-level, we propose a Multi-Scale Attention Module (MSAM), combining a spatial pyramid module with an attention mechanism, when we extract the image features. In addition, we introduce a feature refinement module (FRM) and a 3D attention aggregation module (3D AAM) during cost aggregation so that the network can extract informative features with high representational ability and high-quality channel attention vectors. Finally, we obtain the final disparity through bilinear interpolation and disparity regression. We evaluate our method on the Scene Flow, KITTI 2012 and KITTI 2015 stereo datasets. The experimental results show that our method achieves state-of-theart performance and that every component of our network is effective.
A novel learning‐based end‐to‐end network for stereo matching, named Multi‐path Attention Stereo Matching (MPA‐Net), is introduced in this study. Different from existing methods, the multi‐path attention aggregation module is designed firstly, named MPA, which is a unified structure using three different parallel layers with a respective attention mechanism to extract the multi‐scale informational features. Secondly, the method of cost volume construction, which differs from the traditional stereo matching methods, is extended. And then, the absolute difference between two input features is calculated. Furthermore, a u‐shaped structure with 3D attention gate is selected as the encoder‐decoder module. Specifically, the module is used to fuse the encoding features to their corresponding decoding features under the supervision of the authors' attention gate with skip‐connection, and thus exploit more significant information for matching cost regularisation and disparity prediction. Finally, specific experiments are conducted to evaluate their network on SceneFlow, KITTI2012 and KITTI2015 data sets. The results show that their method achieves a better improvement in disparity maps prediction compared with some existing state‐of‐the‐art methods on KITTI benchmark.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.