发表机构
Singapore University of Technology and Design（新加坡科技设计大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出GUDC，利用ControlNet从稀疏深度合成伪图像，将无引导深度补全转为语义引导，通过多级蒸馏和语义注意力融合提升精度，在KITTI和NYUv2上超越现有方法。
AI 中文摘要
引导式深度补全方法严重依赖RGB图像的质量和对齐，而无引导方法由于缺乏明确的视觉线索，往往精度受限。本文提出了一种新的补全范式——深度到图像合成驱动的生成式无引导深度补全（GUDC），创新性地将先进的2D生成模型与无引导深度补全相结合，无需真实RGB输入即可实现语义感知的深度推断。我们的核心思想是利用ControlNet强大的深度条件生成能力，直接从稀疏深度合成伪图像，从而有效地将原始的无引导设置转化为语义引导设置。为解决深度稀疏性可能导致的图像-深度错位问题，我们提出了一种用于ControlNet微调的多级稠密到稀疏表示蒸馏策略，其中稠密深度特征作为教师信号，为稀疏深度输入蒸馏出一致的结构表示。此外，在伪图像引导的补全过程中，我们提出了伪图像语义注意力融合模块，自适应地从伪图像中提取信息丰富的语义线索，同时抑制伪影（如纹理幻觉）。在KITTI和NYUv2上的大量实验验证了我们的GUDC在精度和鲁棒性上优于现有方法。
英文摘要
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.
