Conference paper
Active-Passive SimStereo - Benchmarking the Cross-Generalization capabilities of deep Learning-based stereo methods
36th Conference on Neural Information Processing Systems (NeurIPS 2022) (New Orleans, 22/11/2022–09/12/2022)
2022
Abstract
In stereo vision, self-similar or bland regions can make it difficult to match patches between two images. Active stereo-based methods mitigate this problem by projecting a pseudo-random pattern on the scene so that each patch of an image pair can be identified without ambiguity. However, the projected pattern significantly alters the appearance of the image. If this pattern acts as a form of adversarial noise, it could negatively impact the performance of deep learning-based methods, which are now the de-facto standard for dense stereo vision. In this paper, we propose the Active-Passive SimStereo dataset and a corresponding benchmark to evaluate the performance gap between passive and active stereo images for stereo matching algorithms. Using the proposed benchmark and an additional ablation study, we show that the feature extraction and matching modules of a selection of twenty selected deep learning-based stereo matching methods generalize to active stereo without a problem. However, the disparity refinement modules of three of the twenty architectures (ACVNet, CascadeStereo, and StereoNet) are negatively affected by the active stereo patterns due to their reliance on the appearance of the input images.
Details
- Title
- Active-Passive SimStereo - Benchmarking the Cross-Generalization capabilities of deep Learning-based stereo methods
- Authors/Creators
- J. LaurentA. AllenL. XuHamid Laga - Murdoch University, Centre for Biosecurity and One HealthF. Boussaid - The University of Western AustraliaMohammed Bennamoun - The University of Western Australia
- Conference
- 36th Conference on Neural Information Processing Systems (NeurIPS 2022) (New Orleans, 22/11/2022–09/12/2022)
- Identifiers
- 991005591969107891
- Murdoch Affiliation
- Murdoch University; Centre for Healthy Ageing; School of Information Technology; Centre for Biosecurity and One Health
- Resource Type
- Conference paper
Metrics
40 File views/ downloads
82 Record Views