LinStereo: Linear-Complexity Global Attention for Multi-Scale Iterative Stereo Matching

• URL: http://arxiv.org/abs/2606.25437v1
• Date: Wed, 24 Jun 2026 05:58:03 GMT
• Title: LinStereo: Linear-Complexity Global Attention for Multi-Scale Iterative Stereo Matching
• Abstract summary: LinStereo is a Position-Aware Linear Attention (PALA) module that replaces local recurrence with global aggregation at linear cost.<n>It achieves state-of-the-art-level accuracy on standard benchmarks and strong cross-domain generalization.<n>It achieves the best overall accuracy with consistent gains 28% lower AbsRel on TartanAir-UW, 26% on SQUID, a real-world underwater dataset.
• Score: 4.952844628691303
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Existing Vision Foundation Model (VFM)-based iterative stereo pipelines under-exploit three information pathways: multi-scale backbone features are collapsed into single-level correlations, geometric priors remain untapped at initialization, and context propagates only locally. These gaps widen under degraded photometric cues, making underwater scenes a stringent generalization test. To address this, we propose LinStereo, built upon Depth Anything V3, whose core is a Position-Aware Linear Attention (PALA) module that replaces local recurrence with global aggregation at linear cost, propagating reliable estimates from well-matched regions into degraded areas while preserving disparity structure. PALA is made effective by two enabling components: Hierarchical Semantic Cost Volumes (HSCV), which supply scale-aligned correlations from the VFM feature hierarchy, and a Depth Prior Initialization (DPI) that converts monocular depth into a metrically calibrated warm start. LinStereo achieves state-of-the-art-level accuracy on standard benchmarks and strong cross-domain generalization, particularly on underwater scene where severe photometric degradation makes stereo matching particularly challenging, attaining the best overall accuracy with consistent gains 28% lower AbsRel on TartanAir-UW, 26% on SQUID, a real-world underwater dataset).

Related papers

• EgoHieraLoc: A Cortically Inspired Hierarchical Segmentation-Guided Framework for Egocentric Visual Query Localization [9.547789251222838]
  Visual query localization (VQL) aims to retrieve and re-localize a queried object in egocentric videos.<n>Human vision handles ambiguity through a hierarchical process.<n>We propose textbfEgoHieraLoc, a unified framework for VQL-2D and VQL-3D.
  arXiv  Detail & Related papers  (2026-08-10T14:31:01Z)
• DGSfM: Depth-Guided Scale-Aware Global Structure-from-Motion [40.127097529332815]
  We propose DGSfM, a depth-aware global SfM pipeline that uses monocular depth maps as a scalable prior.<n>For each image pair, we use a depth-aware relative pose solver to convert scale-ambiguous epipolar constraints into scale-aware relative pose constraints.<n>Experiments on ETH3D and IMC2021 show that DGSfM consistently improves over strong global SfM baselines across sparse and dense matching front-ends.
  arXiv  Detail & Related papers  (2026-07-10T15:18:37Z)
• Focusable Monocular Depth Estimation [49.71122535319022]
  FocusDepth is a prompt-conditioned monocular relative depth estimation framework.<n>It guides depth modeling to focus on target regions via box/text prompts.<n>MSSA's spatial alignment is the key design factor, as disrupting prompt-geometry correspondence increases AbsRel by up to 13.8%.
  arXiv  Detail & Related papers  (2026-05-12T08:30:53Z)
• Progressive Per-Branch Depth Optimization for DEFOM-Stereo and SAM3 Joint Analysis in UAV Forestry Applications [5.266753902938501]
  Per-branch 3D reconstruction is a prerequisite for autonomous UAV-based tree pruning.<n>This paper introduces a progressive pipeline integrating DEFOM-Stereo foundation-model disparity estimation.<n>The proposed pipeline reduces the average per-branch depth standard deviation by 82% while retaining edge fidelity.
  arXiv  Detail & Related papers  (2026-02-24T04:37:18Z)
• StereoAdapter-2: Globally Structure-Consistent Underwater Stereo Depth Estimation [18.410248448681514]
  We propose StereoAdapter-2, which replaces the conventional ConvGRU updater with a novel ConvSS2D operator.<n>We construct UW-StereoDepth-80K, a large-scale synthetic underwater stereo dataset.<n>Our framework achieves state-of-the-art zero-shot performance on underwater benchmarks.
  arXiv  Detail & Related papers  (2026-02-18T22:12:08Z)
• Diving into the Fusion of Monocular Priors for Generalized Stereo Matching [27.15757281613792]
  Recently, stereo matching has progressed by leveraging the unbiased monocular prior from the vision foundation model (VFM) to improve the generalization in ill-posed regions.<n>We propose a binary local ordering map to guide the fusion, which converts the depth map into a binary relative format.<n>We also formulate the final direct fusion of monocular depth to the disparity as a registration problem, where a pixel-wise linear regression module can globally and adaptively align them.
  arXiv  Detail & Related papers  (2025-05-20T14:27:45Z)
• Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model [70.67610495024459]
  Camera-based setups offer a cost-effective option by using stereo depth estimation to generate dense, high-resolution depth maps.<n>Existing omnidirectional stereo matching approaches achieve only limited depth accuracy across diverse environments.<n>We present DFI-OmniStereo, a novel omnidirectional stereo matching method that leverages a large-scale pre-trained foundation model for relative monocular depth estimation.
  arXiv  Detail & Related papers  (2025-03-30T16:24:22Z)
• FoundationStereo: Zero-Shot Stereo Matching [50.79202911274819]
  FoundationStereo is a foundation model for stereo depth estimation.<n>We first construct a large-scale (1M stereo pairs) synthetic training dataset.<n>We then design a number of network architecture components to enhance scalability.
  arXiv  Detail & Related papers  (2025-01-17T01:01:44Z)
• DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation [17.99904937160487]
  DCPI-Depth is a framework that incorporates all these innovative components and couples two bidirectional and collaborative streams.<n>It achieves state-of-the-art performance and generalizability across multiple public datasets, outperforming all existing prior arts.
  arXiv  Detail & Related papers  (2024-05-27T08:55:17Z)
• DETR Doesn't Need Multi-Scale or Locality Design [69.56292005230185]
  This paper presents an improved DETR detector that maintains a "plain" nature. It uses a single-scale feature map and global cross-attention calculations without specific locality constraints. We show that two simple technologies are surprisingly effective within a plain design to compensate for the lack of multi-scale feature maps and locality constraints.
  arXiv  Detail & Related papers  (2023-08-03T17:59:04Z)
• On Robust Cross-View Consistency in Self-Supervised Monocular Depth Estimation [56.97699793236174]
  We study two kinds of robust cross-view consistency in this paper. We exploit the temporal coherence in both depth feature space and 3D voxel space for self-supervised monocular depth estimation. Experimental results on several outdoor benchmarks show that our method outperforms current state-of-the-art techniques.
  arXiv  Detail & Related papers  (2022-09-19T03:46:13Z)
• Non-local Recurrent Regularization Networks for Multi-view Stereo [108.17325696835542]
  In deep multi-view stereo networks, cost regularization is crucial to achieve accurate depth estimation. We propose a novel non-local recurrent regularization network for multi-view stereo, named NR2-Net. Our method achieves state-of-the-art reconstruction results on both DTU and Tanks and Temples datasets.
  arXiv  Detail & Related papers  (2021-10-13T01:43:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.