EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

BRD-Stereo: Bilateral Reliability Distillation for Lightweight Zero-Shot Stereo Matching

Yitong Zhang, Bobin Yao

Symmetry · 29.10.2026 · 18 (11) · 1819

DOI: 10.3390/sym18111819 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Symmetry, online erschienen am 29.10.2026. Verfasst von Yitong Zhang und 1 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Basic Research Program of Shaanxi.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stereo foundation models provide strong cross-domain generalization but remain computationally expensive for real-time and resource-constrained deployment. We propose BRD-Stereo, a bilateral reliability distillation framework that transfers foundation-model stereo knowledge to a lightweight BANet-style student. BRD-Stereo treats reciprocal left–right correspondence after geometric warping as an exchange symmetry and uses localized symmetry violations from occlusion or mismatching to reduce teacher reliability. BRD-Stereo models the spatial reliability of teacher supervision and aligns different forms of teacher knowledge with the student’s detail/smooth aggregation structure. The detail branch receives boundary- and gradient-aware supervision, while the smooth branch is guided by low-frequency disparity consistency and local plane regularization. Reliability cues from left–right consistency, image disparity edge agreement, local regularity, occlusion-aware validity, and available teacher confidence weight the pseudo-label objectives. Sliced score distribution distillation transfers a locally softened disparity target around the teacher prediction without requiring full teacher cost volume supervision. Experiments on Scene Flow, KITTI, ETH3D, and Middlebury show consistent gains over plain pseudo-label distillation, with the largest improvements concentrated in boundary, occlusion, weak-texture, and large-disparity regions while preserving BANet-scale inference efficiency.

Förderung

  • Natural Science Basic Research Program of Shaanxi

Zitieren

APA

Zhang, Y., & Yao, B. (2026). BRD-Stereo: Bilateral Reliability Distillation for Lightweight Zero-Shot Stereo Matching. Symmetry, 18(11), 1819. https://doi.org/10.3390/sym18111819

Vancouver

Zhang Y, Yao B. BRD-Stereo: Bilateral Reliability Distillation for Lightweight Zero-Shot Stereo Matching. Symmetry. 2026;18(11):1819. doi:10.3390/sym18111819

BibTeX
@article{Zhang2026,
  author = {Zhang, Yitong and Yao, Bobin},
  title = {BRD-Stereo: Bilateral Reliability Distillation for Lightweight Zero-Shot Stereo Matching},
  journal = {Symmetry},
  year = {2026},
  volume = {18},
  number = {11},
  pages = {1819},
  doi = {10.3390/sym18111819}
}
Kostenlos starten →