Multi-Hypothesis Spatial Foundation Model: Rethinking and Decoupling Depth Ambiguity via Laplacian Visual Prompting

Xiaohao Xu, Feng Xue, Xiang Li, Haowei Li, Shusheng Yang, Tianyi Zhang, Matthew Johnson-Roberson, Xiaonan Huang

Keywords: Spatial Ambiguity, Visual Prompting, Depth Estimation, Foundation Model

TL;DR: Unlocking hidden spatial knowledge (i.e., multi-layer depth) in ambiguous 3D scenes via Laplacian Visual Prompting.

Abstract:

Depth ambiguity is a fundamental challenge in spatial scene understanding, especially in transparent scenes where single-depth estimates fail to capture full 3D structure. Existing models, limited to deterministic predictions, overlook real-world multi-layer depth. To address this, we introduce a paradigm shift from single-prediction to multi-hypothesis spatial foundation models.
We first present MD-3k, a benchmark exposing depth biases in expert and foundational models through multi-layer spatial relationship labels and new metrics.
To resolve depth ambiguity, we propose Laplacian Visual Prompting (LVP), a training-free spectral prompting technique that extracts hidden depth from pre-trained models via Laplacian-transformed RGB inputs. By integrating LVP-inferred depth with standard RGB-based estimates, our approach elicits multi-layer depth without model retraining.
Extensive experiments validate the effectiveness of LVP in zero-shot multi-layer depth estimation, unlocking more robust and comprehensive geometry-conditioned visual generation, 3D-grounded spatial reasoning, and temporally consistent video-level depth inference. Our benchmark and code will be available at https://github.com/Xiaohao-Xu/Ambiguity-in-Space.

Submission Number: 109