用分形结构实现更稳定的单目深度连续生成。
Fractal Autoregressive Depth Estimation with Continuous Token Diffusion
- 分层递归生成，从粗到细逐步预测深度图
- 在KITTI和Cityscapes上达到新最优，误差低于0.123
- 适合需要高精度与可靠性的自动驾驶场景
单目深度估计可受益于自回归（AR）生成，但直接建模受限于RGB与深度之间的模态差异、逐像素生成效率低以及连续深度预测的不稳定性。本文提出一种分形视觉自回归扩散框架，将深度估计重构为粗到细的多尺度自回归生成过程。引入VCFR模块融合多尺度图像特征与当前深度预测，增强跨模态条件建模；采用条件去噪扩散损失，直接在连续空间建模深度分布，缓解离散量化带来的误差。为提升计算效率，将各尺度生成器组织成分形递归结构，复用基础视觉自回归单元构成自相似层级。进一步设计不确定性感知的鲁棒共识聚合方案，提升多样本推理融合稳定性，并提供像素级可靠性估计。在标准基准测试上验证了方法的有效性，性能显著优于现有方法。
原文摘要 · Abstract (English)
Monocular depth estimation can benefit from autoregressive (AR) generation, but direct AR modeling is hindered by the modality gap between RGB and depth, inefficient pixel-wise generation, and instability in continuous depth prediction. We propose a Fractal Visual Autoregressive Diffusion framework that reformulates depth estimation as a coarse-to-fine, next-scale autoregressive generation process. A VCFR module fuses multi-scale image features with current depth predictions to improve cross-modal conditioning, while a conditional denoising diffusion loss models depth distributions directly in continuous space and mitigates errors caused by discrete quantization. To improve computational efficiency, we organize the scale-wise generators into a fractal recursive architecture, reusing a base visual AR unit in a self-similar hierarchy. We further introduce an uncertainty-aware robust consensus aggregation scheme for multi-sample inference to improve fusion stability and provide a practical pixel-wise reliability estimate. Experiments on standard benchmarks demonstrate strong performance and validate the effectiveness of the proposed design.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。