Enhancing Stereo Matching Domain Generalization With Adversarial Domain Alignment
- 13 May 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23 (4) , 9400-9411
Abstract
Recently, state-of-the-art stereo-matching networks trained on large-scale synthetic data have shown remarkable performance. However, their capacity to extrapolate effectively to unseen real-world data, i.e. different domains, remains a challenge. The major difficulty resides in the unforeseeable domain gap when generalizing from synthetic data to real-world data. In this paper, we introduce ADASM, an approach using adversarial domain alignment, designed to enhance the robustness and generalization of stereo-matching networks. It mainly consists of two modules: an end-to-end robustness optimizer and a domain-invariant feature learner. First, we adapt adversarial training into the stereo-matching task to reduce models' sensitivity to the perturbation in real-world samples. By introducing worst cases into the training space, we take unseen data into account and achieve robust disparity estimation for the end-to-end model. Then, via simulating the real-world noise with gradient-based perturbation, we construct a fictitious domain, which is taken as a referential distribution of the real-world noisy data, for further domain alignment. Specifically, we propose to utilize Maximum Mean Discrepancy to realize domain regularization between the original domain and the fictitious one. Finally, we fuse all aforementioned objectives and propose a unified, simple but effective loss function that can be adapted to all stereo-matching networks. The extensive experiments show that our method achieves a superior disparity estimation performance on various real-world benchmarks, including KITTI, Middlebury, and DrivingStereo. More importantly, ADASM obtains competitive or even better performance than the fine-tuning strategy, revealing its fine-tuning-free character.Keywords
Funding Information
- National Natural Science Foundation of China (62502075)
- General Program of the Sichuan Provincial Natural Science Foundation (2026NSFSC0431)
This publication has 49 references indexed in Scilit:
- Improving Domain Generalization in Self-supervised Monocular Depth Estimation via Stabilized Adversarial Training2024
- Learning Representations from Foundation Models for Domain Generalized Stereo Matching2024
- Bearing fault diagnosis with intermediate domain based Layered Maximum Mean Discrepancy: A new transfer learning approachEngineering Applications of Artificial Intelligence, 2021
- Domain-Invariant Stereo Matching Networks2020
- Object Scene FlowISPRS Journal of Photogrammetry and Remote Sensing, 2018
- Overcoming catastrophic forgetting in neural networksProceedings of the National Academy of Sciences, 2017
- High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth2014
- Reproducing Kernel Hilbert Spaces in Probability and Statistics2004
- Vision for mobile robot navigation: a surveyIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- Non-parametric local transforms for computing visual correspondence1994