The paper audits the confidence outputs of seven feed‑forward 3D reconstruction backbones across 13 datasets, evaluating four properties: error ranking, average error‑to‑uncertainty ratio, slope of this ratio, and coverage of the implied error distribution. While confidence ranks errors well, the decoded uncertainty is consistently too small—off by at least 2.4× on median cases—and worsens with higher confidence. A post‑hoc power‑law fit per backbone‑dataset pair improves all four metrics at the dataset level, reducing the median error by 1.35×, but fails to correct coverage for many held‑out scenes, indicating the models lack the correct error scale and distribution shape.
By Nanxing Nick Deng, Qing Cheng, Niclas Zeller, Daniel Cremers
arXiv:2610.07883v1 Announce Type: new
Abstract: A correctly reconstructed distant point appears uncertain even when a frozen 3D model processes equivalent inputs because its output frame rotates frac...
By Sammam Mahdi, Fariha Binta Salim, Rakin Bin Rabbani, Aniqua Nusrat Zereen
The paper introduces a counterfactual benchmark to evaluate whether vision models understand physical constraints or merely learn rendering shortcuts. Using Mitsuba 3, it generates 5,500 synthetic images across 500 scene families, each with a clean image and ten violations (shadows, reflections, support, surface response, occlusion). Diagnostic baselines (LLaVA‑1.5‑7B, Qwen2.5‑VL‑7B, InternVL3.5‑8B) achieve 64.0–67.8 % macro‑F1 on a held‑out synthetic test set, but performance drops to 40.8 % for LLaVA‑1.5‑7B under an intervention shift, indicating reliance on rendering cues rather than true physical reasoning.
By M. Moein Esfahani, Sepehr Salem, Mohammed Alser, Vince Calhoun
arXiv:2608. 07579v1 Announce Type: cross Abstract: The AI City Challenge 2026 Track 1 evaluates multi-camera 3D perception in large indoor warehouses under a synthetic-to-real (Sim2Real) setting; depth is available only for training and validation, so inference is RGB-only.
By Abdullah Naeem, Anav Katwal, Ayon Dey, Noman Khan, Md Tamjidul Hoque
A surveillance camera is an image sensor whose silent physical degradation invalidates every downstream consumer of its data. In-situ integrity alarms for such vision sensors require low false-alarm rates, bounded computation, and diagnosable behavior under nuisance illumination changes.
arXiv:2610.35806v1 Announce Type: new
Abstract: Feed-forward 3D reconstruction models such as VGGT regress cameras and geometry in a single forward pass, but they report no uncertainty for the camera...
By Adityan Thirumani Srinivasan, David Nordstr\"om
arXiv:2608.29705v1 Announce Type: cross
Abstract: Feed-forward 3D reconstruction models emit a per-pixel confidence that downstream systems read as a reliability signal. It is trained as a loss weigh...
By Nanxing Nick Deng, Qing Cheng, Niclas Zeller, Daniel Cremers
ObjectTwin introduces a scan‑grounded protocol for measuring how well generated 3D assets match captured scenes. By fixing the scene, cameras, and depth data and only swapping the object representation, the method quantifies the asset correspondence gap—how much observation agreement is lost compared to the scan‑derived geometry. Experiments on two indoor RGB‑D datasets and two generation backends show that anchored generated assets lose 0.302–0.497 MaskIoU, with scene‑level estimates consistently positive but generator preferences varying across domains.
By Zhenghe Guo
arXiv:2606. 16479v1 Announce Type: cross Abstract: Visual Geometry Grounded Transformer (VGGT) has already attracted a great deal of attention in a short period of time, not least due to the Best Paper Award at CVPR-2025.
By Markus Hillemann, Robert Langend\"orfer, Steven Landgraf, Markus Ulrich
arXiv:2607. 22034v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly deployed on consumer hardware where input images are degraded by compression, camera shake, and poor lighting.
By M M Asif Ferdous
arXiv:2608. 05670v1 Announce Type: new Abstract: A model's agreement across perturbed inputs is used both as a label-free reliability signal and as a self-training target, on the premise that agreement tracks correctness.
By Rasul Khanbayov, Hasan Kurban
The paper introduces a no-reference point cloud quality assessment method that leverages a complexity‑calibrated reconstruction discrepancy. By reconstructing a distorted point cloud and comparing successive reconstructions with varying smoothing, the approach isolates a proxy for structural complexity, which is then used to calibrate the initial reconstruction error. The calibrated geometric and projected‑view discrepancies, measured with frozen pretrained perceptual representations, are fused into a final degradation score without requiring any PCQA‑specific training.
By Haocheng Wu, Xiangjie Sui, Xin Sun