arXiv:2607.13682v4 Announce Type: replace-cross Abstract: Does an uncertainty map identify where a reconstruction is wrong? In sparse-view computed tomography (CT), we find a sharp gap between whole-...
VOIM (Voxel‑Grounded Online Instance Manager) is a training‑free system that builds open‑vocabulary 3D instance maps from RGB‑D or monocular RGB input by deferring label and instance decisions until sufficient soft evidence accumulates per voxel across views. Across four perception configurations on ScanNet++, VOIM outperforms the strongest online RGB‑D system, OVO‑SLAM, by 4.8–11.7 mIoU, and achieves 44.07 mIoU under a like‑for‑like protocol, winning all ten scenes. The method also runs unchanged on monocular RGB, matching baseline performance on Replica, and produces exportable occupancy grids that support free‑form instance queries.
arXiv:2607. 13682v2 Announce Type: cross Abstract: Radiative Gaussian splatting reconstructs sparse-view CT fast and accurately, and recent work attaches per-Gaussian posteriors to yield per-voxel uncertainty maps.
arXiv:2606. 18451v1 Announce Type: new Abstract: Single-image-to-3D generators are improving quickly, but there is no agreed, human-free way to tell whether one generated mesh is better than another.
arXiv:2609.10307v1 Announce Type: cross Abstract: 3D Gaussian Splatting (3DGS) renders novel views in real time, but an uncertainty heatmap does not certify that a rendered view meets a certain predi...
The paper argues that internal self-consistency checks cannot guarantee the accuracy of photogrammetric reconstructions, a limitation that is structural rather than a tuning issue. It introduces a track‑leakage‑free hold‑out protocol that withholds a deterministic subset of images and tests each against only 3D points supported by at least two retained images, ensuring no view is evaluated against the structure it helped create. Experiments on diverse datasets show that while the protocol is well‑posed, it saturates at a confidence score of 1.00 and fails to detect coherent distortion, missing large errors that can reach over 100 m. whyItMatters":"The study highlights that hold‑out self‑validation scores, increasingly used as quality evidence for metric deliverables, may be misleading and cannot replace external survey validation."
The paper introduces a 2D Gaussian Splatting pipeline that renders a dominant-eye RGB image and depth proxy, then reprojects and selectively patches the affiliated eye to reduce redundant work. By reusing alpha-blending weights and generating adaptive regions of interest, the method cuts sequential binocular rendering time by 15.5% to 28.8% and GPU memory by 6% to 11% on several datasets, with minimal quality loss. It demonstrates a practical efficiency‑quality trade‑off for static‑scene stereo rendering and suggests further evaluation on dynamic scenes and VR hardware.
arXiv:2607.02554v2 Announce Type: replace Abstract: Sparse-view neural reconstruction in outdoor driving is challenging due to narrow forward-facing trajectories and limited multi-view overlap, and m...
3D Gaussian Splatting (3DGS) renders novel views in real time, but an uncertainty heatmap does not certify that a rendered view meets a certain prediction coverage. We treat novel-view synthesis as st...
The paper investigates why post‑hoc saliency maps, such as Grad‑CAM, shift when input images are rotated, even if the model’s prediction remains unchanged. By measuring equivariance at each stage of the CAM operator, the authors find that the instability originates from the spatial activation tensor rather than the channel weights, and that a training‑free wrapper called EquiGrad‑CAM can align and average saliency maps across rotated views to significantly improve rotation equivariance. Experiments on ImageNet, PatchCamelyon, and RESISC45 demonstrate that EquiGrad‑CAM outperforms rotation‑augmented training and enhances zero‑shot CLIP explanations.
Syn2RealTrack addresses the synthetic‑to‑real gap in multi‑camera 3D perception for warehouses by decomposing it into three distinct issues: camera calibration, object shape prior, and known object census. The pipeline corrects lens distortion from images, fuses detections with a visibility‑weighted part‑based descriptor, measures person height directly from calibration, and uses a closed‑world cardinality prior with a causal filter to eliminate phantom boxes. These local remedies allow the system to adapt without retraining a feature extractor, achieving a 3D HOTA of 52.0118% on the AI City Challenge 2026 Track 1.
CALIPER is a new benchmark that tests whether pretrained visual encoders can infer physical properties such as mass and friction from images. The test involves striking an object twice at known speeds, showing a third strike only up to contact, and asking a linear readout on frozen features to predict how far the object slides. Results show that in clean, fixed‑camera scenes all representations perform similarly, but when camera, lighting, and clutter are varied, only encoders that truly infer physics—like V‑JEPA 2—maintain performance, while random or raw pixel representations fail.