arXiv:2610.08756v1 Announce Type: new Abstract: The same Gaussian of a 3D Gaussian Splatting model is seen from many views, and these views do not always agree on the class it belongs to. The Gaussia...
The paper investigates how much semantic information is lost when frozen foundation models are combined for few‑shot 3D segmentation. By varying the number of retained semantic alternatives before fusion, the authors show that keeping the full distribution of class scores yields higher harmonic‑mean IoU than collapsing to a single class. Experiments on ScanNet200 and ScanNet++ confirm that full‑distribution fusion consistently outperforms top‑1 and other operators, and that most useful information is recovered by retaining a compact set of plausible alternatives.
arXiv:2608.28216v1 Announce Type: new Abstract: Locating a specific object instance in a cluttered scene using a single reference image and a short description, and reporting when that instance is ab...
The paper introduces an open‑vocabulary 3D object detection pipeline that uses a promptable segmentation model (SAM3) to generate instance masks from six surround‑view cameras. These masks are converted into metric 3D boxes, achieving up to 0.413 mAP/0.555 NDS without any training when supervised box geometry is borrowed at inference. The approach also improves a supervised LiDAR‑only detector by 0.034 mAP through a camera‑witness rule, demonstrating that measurement precision, not 2D detection, limits performance.
arXiv:2606. 04364v1 Announce Type: cross Abstract: Concept bottleneck models (CBMs) predict a layer of human-named attributes before predicting a class, which makes their decisions auditable.
The paper introduces SphereTrust, a method that uses frozen self‑supervised hyperspherical features to evaluate and rank candidate masks produced by foundation segmenters like SAM. By measuring angular contrast, foreground coverage, and image‑frame contact, SphereTrust can select high‑quality masks in 0.55 s per image and outperforms existing baselines on multiple segmentation tasks. The selected masks are then used as priors to train student models, improving performance on several benchmark datasets.
arXiv:2606.18623v2 Announce Type: replace Abstract: Gaussian segmentation is usually posed as transferring object knowledge from 2D foundation models into a 3D representation. This leaves a fundament...
arXiv:2604. 27720v2 Announce Type: replace Abstract: Vision-language models (VLMs) are increasingly applied to medical visual question answering (Med-VQA), yet whether they can \emph{localize} the evidence behind their answers---a prerequisite for clinical auditability---is poorly characterized.
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.
Gauss What You Need: Compact Gaussian Splatting Across Scene Scales introduces TangoGS, a method that automatically selects the number of Gaussian primitives for 3D Gaussian Splatting by combining capture-derived model sizing with training-based adaptation. The approach first estimates a learning allowance based on the capture’s total pixels, then adjusts the number of Gaussians during training according to reconstruction quality. On standard benchmarks, TangoGS matches the best baseline’s PSNR while using 48% fewer Gaussians, and on larger captures it scales automatically to achieve the highest mean PSNR with 2.3× more Gaussians.
arXiv:2609.18898v1 Announce Type: new Abstract: Training-free weighted aggregation is widely used to lift 2D semantic features onto 3D Gaussians for open-vocabulary scene understanding, yet its theor...
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.