arXiv:2607. 07962v1 Announce Type: cross Abstract: Inferring latent physical properties from sensory observations is a fundamental challenge in machine perception.
RAFT-DVC is a resolution‑aware family of recurrent all‑pairs field transform (RAFT) based digital volume correlation (DVC) solvers that use encoder downsampling factors of 2, 4, and 8. The solvers localize displacement to about 0.017 feature‑grid voxels, with raw‑volume error scaling roughly as 0.017 s voxels, and exhibit complementary operating regimes determined by displacement reach and volumetric‑texture compatibility. Synthetic benchmarks show comparable performance to tuned classical DVC for fine‑texture, small‑to‑moderate displacements, while outperforming it for coarse‑texture, large‑displacement scenarios; additional tests on confocal and micro‑CT images confirm the importance of matching solver regimes to deformation magnitude and texture, and demonstrate cross‑texture transfer and improved accuracy after correcting sampler geometry.
arXiv:2609. 03378v1 Announce Type: new Abstract: Depth can resolve appearance ambiguity in RGB-D salient object detection (SOD), yet sensor depth is not uniformly reliable.
arXiv:2604.12251v2 Announce Type: replace Abstract: 3D Gaussian Splatting (3DGS) delivers high-fidelity real-time rendering but suffers from geometric and photometric degradations under sparse-view c...
XCalib is an unsupervised dense registration framework that aligns thermal and visible video streams by optimizing virtual pinhole camera parameters and predicted monocular depth, thereby restricting spatial displacements to physically valid projection geometries. It introduces a novel registration paradigm using camera parameterization as an implicit regularizer, a robust Normalized Edges Correlation (NEC) metric for cross‑spectral alignment, and demonstrates superior temporal stability and alignment accuracy on public ADAS datasets compared to unconstrained dense flow baselines.
arXiv:2606. 29600v1 Announce Type: cross Abstract: A faithful 3D world representation should account for layered geometry, where a single camera ray may contain multiple visible and geometrically valid surfaces.
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.
arXiv:2609.24531v1 Announce Type: new Abstract: Thermography plays a vital role in military and broader thermal analysis applications. Recent progress in 3D thermal reconstruction has extended temper...
arXiv:2609.36929v1 Announce Type: new Abstract: Recent event-based depth estimation methods successfully transfer geometric priors from vision foundation models via cross-modal distillation. However,...
arXiv:2608.23206v1 Announce Type: new Abstract: We study spherical occupancy profiles-the ray-wise occupancy probability profiles P(r) = T(r) o(r) distilled from multi-view 3D Gaussian reconstruction...
The paper presents a close‑range photogrammetry workflow using Structure‑from‑Motion and Multi‑View Stereo to generate high‑resolution 3D point clouds of rubberised concrete. By capturing images with a Canon DSLR and an iPhone 16, the authors achieved sub‑millimetre reconstruction accuracy, outperforming traditional LiDAR for fine‑scale defect analysis. An RGB‑guided crack extraction method and deformation analysis further demonstrate the method’s utility for detailed surface monitoring and material performance evaluation.
arXiv:2608.30521v1 Announce Type: new Abstract: Conventional algebraic triangulation solves 3D human pose estimation (HPE) from multi-view 2D keypoints. The typical approach, decoding 2D keypoints fr...