Metraipolis › archive › dc_a78cd049ef3c

Depth-to-Image Synthesis-Driven Generative Unguided Depth Completion — arXiv:2609.06007

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.06007
License (as archived)
CC0-1.0
Archived
2026-10-16T15:06:46.462Z
SHA-256 of the archived text
835223f5102cced842481150dc5ee738cd2d8dfc1157b3262a70329c12c39c76
Stable URL of this page
https://metraipolis.com/documento/dc_a78cd049ef3c

How to cite

Depth-to-Image Synthesis-Driven Generative Unguided Depth Completion — arXiv:2609.06007 — arXiv (en), https://arxiv.org/abs/2609.06007 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a78cd049ef3c (Metraipolis archive dc_a78cd049ef3c, sha256 835223f5102cced842481150dc5ee738cd2d8dfc1157b3262a70329c12c39c76)
BibTeX
@misc{dca78cd049ef3c,
  title = {Depth-to-Image Synthesis-Driven Generative Unguided Depth Completion — arXiv:2609.06007},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06007},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a78cd049ef3c; sha256 835223f5102cced842481150dc5ee738cd2d8dfc1157b3262a70329c12c39c76},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06007. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Guided depth completion methods heavily depend on RGB quality and alignment, while unguided ones often suffer from limited precision due to the absence of explicit visual cues. In this paper, we present Depth-to-Image Synthesis-Driven Generative Unguided Depth Completion (GUDC), a new completion paradigm that innovatively bridges advanced 2D generative models with unguided depth completion, enabling semantics-aware depth inference without real RGB inputs. Our key idea is to exploit ControlNet's powerful depth-conditioned generation capability to synthesize pseudo-images directly from sparse depth, effectively converting the original unguided setting into a semantics-guided one. To address the potential image-depth misalignment caused by depth sparsity, we propose a multi-level dense-to-sparse representation distillation strategy for ControlNet fine-tuning, where dense-depth features act as teacher signals to distill consistent structural representations for sparse-depth inputs. Furthermore, during pseudo-image-guided completion, we propose a pseudo-image semantic attention fusion module to adaptively extract informative semantic cues from pseudo-images while suppressing artifacts (e.g., texture hallucinations). Extensive experiments on KITTI and NYUv2 validate that our GUDC achieves superior accuracy and robustness over existing methods.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.06007 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T14:49:57.035Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).