Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Prior
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (4) · 3877-3885
DOI: 10.1609/aaai.v39i4.32405 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Lee Hyoseok und 3 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref).
Abstract (Originalsprache)
Depth completion, predicting dense depth maps from sparse depth measurements, is an ill-posed problem requiring prior knowledge. Recent methods adopt learning-based approaches to implicitly capture priors, but the priors primarily fit in-domain data and do not generalize well to out-of-domain scenarios. To address this, we propose a zero-shot depth completion method composed of an affine-invariant depth diffusion model and test-time alignment. We use pre-trained depth diffusion models as depth prior knowledge, which implicitly understand how to fill in depth for scenes. Our approach aligns the affine-invariant depth prior with metric-scale sparse measurements, enforcing them as hard constraints via an optimization loop at test-time. Our zero-shot depth completion method demonstrates generalization across various domain datasets, achieving up to a 21% average performance improvement over the previous state-of-the-art methods while enhancing spatial understanding by sharpening scene details. We demonstrate that aligning a monocular affine-invariant depth prior with sparse metric measurements is a sufficient strategy to achieve domain-generalizable depth completion without relying on extensive training datasets.
Zitieren
APA
Hyoseok, L., Kim, K. S., Byung-Ki, K., & Oh, T. (2025). Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Prior. Proceedings of the AAAI Conference on Artificial Intelligence, 39(4), 3877-3885. https://doi.org/10.1609/aaai.v39i4.32405
Vancouver
Hyoseok L, Kim KS, Byung-Ki K, Oh T. Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Prior. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(4):3877-3885. doi:10.1609/aaai.v39i4.32405
BibTeX
@article{Hyoseok2025,
author = {Hyoseok, Lee and Kim, Kyeong Seon and Byung-Ki, Kwon and Oh, Tae-Hyun},
title = {Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Prior},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {4},
pages = {3877--3885},
doi = {10.1609/aaai.v39i4.32405}
}
EUVIMED