Photorealistic Phantom Roads in Real Scenes: Disentangling 3D Hallucinations from Physical Geometry

• URL: http://arxiv.org/abs/2512.15423v1
• Date: Wed, 17 Dec 2025 13:14:37 GMT
• Title: Photorealistic Phantom Roads in Real Scenes: Disentangling 3D Hallucinations from Physical Geometry
• Abstract summary: This paper introduces the first end-to-end framework to probe, quantify, and tame this unquantified safety risk.<n>To probe, we present 3D-Mirage, the first benchmark of real-world illusions with precise planar-region annotations and context-restricted crops.<n>To quantify, we propose a Laplacian-based evaluation framework with two metrics: the Deviation Composite Score (DCS) for spurious non-planarity and the Confusion Composite Score (CCS) for contextual instability.
• Score: 17.571800971505787
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Monocular depth foundation models achieve remarkable generalization by learning large-scale semantic priors, but this creates a critical vulnerability: they hallucinate illusory 3D structures from geometrically planar but perceptually ambiguous inputs. We term this failure the 3D Mirage. This paper introduces the first end-to-end framework to probe, quantify, and tame this unquantified safety risk. To probe, we present 3D-Mirage, the first benchmark of real-world illusions (e.g., street art) with precise planar-region annotations and context-restricted crops. To quantify, we propose a Laplacian-based evaluation framework with two metrics: the Deviation Composite Score (DCS) for spurious non-planarity and the Confusion Composite Score (CCS) for contextual instability. To tame this failure, we introduce Grounded Self-Distillation, a parameter-efficient strategy that surgically enforces planarity on illusion ROIs while using a frozen teacher to preserve background knowledge, thus avoiding catastrophic forgetting. Our work provides the essential tools to diagnose and mitigate this phenomenon, urging a necessary shift in MDE evaluation from pixel-wise accuracy to structural and contextual robustness. Our code and benchmark will be publicly available to foster this exciting research direction.

Related papers

• Reliability-Aware 3D Geometric Injection for Universal Person Re-identification [47.874819213762926]
  We propose a Universal Monocular 3D-Enhanced ReID framework driven by a Consistency-Aware Reliability Gate and Dual-Stream Residual Fusion.<n>Our method improves challenging, structure-sensitive scenarios while preserving competitive performance on clean domains.
  arXiv  Detail & Related papers  (2026-07-21T08:54:25Z)
• Robust 3D Alignment of Generative Reconstructions via Partial Monocular Observations [46.983222121901235]
  Generative-to-Partial Monocular Observational Alignment is a critical but under-explored challenge in computer vision.<n>We propose a training-free and interpretable geometric alignment framework that grounds generative 3D priors via a 3D similarity transformation.<n>Experiments demonstrate that our method achieves stable and accurate registration, substantially outperforming both classical geometric pipelines and state-of-the-art learning-based baselines.
  arXiv  Detail & Related papers  (2026-07-01T06:30:47Z)
• LESV: Language Embedded Sparse Voxel Fusion for Open-Vocabulary 3D Scene Understanding [9.377694035678948]
  We present a novel framework that leverages Sparse Voxel Rasterization (SVRaster) as a structured, disjoint geometry representation.<n>This enables a deterministic, confidence-aware feature registration process and suppresses the semantic bleeding artifact common in 3DGS.<n>Our approach achieves state-of-the-art performance on Open Vocabulary 3D Object Retrieval and Point Cloud Understanding benchmarks.
  arXiv  Detail & Related papers  (2026-04-01T20:48:06Z)
• StableGS: A Floater-Free Framework for 3D Gaussian Splatting [9.935869165752283]
  3D Gaussian Splatting (3DGS) reconstructions are plagued by stubborn floater" artifacts that degrade their geometric and visual fidelity.<n>We propose StableGS, a novel framework that decouples geometric regularization from final appearance rendering.<n> Experiments on multiple benchmarks show StableGS not only eliminates floaters but also resolves the common blur-artifact trade-off.
  arXiv  Detail & Related papers  (2025-03-24T09:02:51Z)
• Learning A Zero-shot Occupancy Network from Vision Foundation Models via Self-supervised Adaptation [41.98740330990215]
  This work proposes a novel approach that bridges 2D vision foundation models with 3D tasks.<n>We leverage the zero-shot capabilities of vision-language models for image semantics.<n>We project the semantics into 3D space using the reconstructed metric depth, thereby providing 3D supervision.
  arXiv  Detail & Related papers  (2025-03-10T09:54:40Z)
• PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting [54.7468067660037]
  PF3plat sets a new state-of-the-art across all benchmarks, supported by comprehensive ablation studies validating our design choices.<n>Our framework capitalizes on fast speed, scalability, and high-quality 3D reconstruction and view synthesis capabilities of 3DGS.
  arXiv  Detail & Related papers  (2024-10-29T15:28:15Z)
• Toward Availability Attacks in 3D Point Clouds [28.496421433836908]
  We show that extending 2D availability attacks directly to 3D point clouds under distance regularization is susceptible to the degeneracy. We propose a novel Feature Collision Error-Minimization (FC-EM) method, which creates additional shortcuts in the feature space. Experiments on typical point cloud datasets, 3D intracranial aneurysm medical dataset, and 3D face dataset verify the superiority and practicality of our approach.
  arXiv  Detail & Related papers  (2024-06-26T08:13:30Z)
• FILP-3D: Enhancing 3D Few-shot Class-incremental Learning with Pre-trained Vision-Language Models [59.13757801286343]
  Few-shot class-incremental learning aims to mitigate the catastrophic forgetting issue when a model is incrementally trained on limited data.<n>We introduce the FILP-3D framework with two novel components: the Redundant Feature Eliminator (RFE) for feature space misalignment and the Spatial Noise Compensator (SNC) for significant noise.
  arXiv  Detail & Related papers  (2023-12-28T14:52:07Z)
• RadOcc: Learning Cross-Modality Occupancy Knowledge through Rendering Assisted Distillation [50.35403070279804]
  3D occupancy prediction is an emerging task that aims to estimate the occupancy states and semantics of 3D scenes using multi-view images. We propose RadOcc, a Rendering assisted distillation paradigm for 3D Occupancy prediction.
  arXiv  Detail & Related papers  (2023-12-19T03:39:56Z)
• Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimation [70.32536356351706]
  We introduce MRP-Net that constitutes a common deep network backbone with two output heads subscribing to two diverse configurations. We derive suitable measures to quantify prediction uncertainty at both pose and joint level. We present a comprehensive evaluation of the proposed approach and demonstrate state-of-the-art performance on benchmark datasets.
  arXiv  Detail & Related papers  (2022-03-29T07:14:58Z)
• PONet: Robust 3D Human Pose Estimation via Learning Orientations Only [116.1502793612437]
  We propose a novel Pose Orientation Net (PONet) that is able to robustly estimate 3D pose by learning orientations only. PONet estimates the 3D orientation of these limbs by taking advantage of the local image evidence to recover the 3D pose. We evaluate our method on multiple datasets, including Human3.6M, MPII, MPI-INF-3DHP, and 3DPW.
  arXiv  Detail & Related papers  (2021-12-21T12:48:48Z)
• MonoRUn: Monocular 3D Object Detection by Reconstruction and Uncertainty Propagation [4.202461384355329]
  We propose MonoRUn, a novel 3D object detection framework that learns dense correspondences and geometry in a self-supervised manner. Our proposed approach outperforms current state-of-the-art methods on KITTI benchmark.
  arXiv  Detail & Related papers  (2021-03-23T15:03:08Z)
• 3D Dense Geometry-Guided Facial Expression Synthesis by Adversarial Learning [54.24887282693925]
  We propose a novel framework to exploit 3D dense (depth and surface normals) information for expression manipulation. We use an off-the-shelf state-of-the-art 3D reconstruction model to estimate the depth and create a large-scale RGB-Depth dataset. Our experiments demonstrate that the proposed method outperforms the competitive baseline and existing arts by a large margin.
  arXiv  Detail & Related papers  (2020-09-30T17:12:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.