arXiv Computer Vision

Spheriverse: 3D Scene Understanding from Spherical Observations in the Wild

arXiv Computer Vision
Sep 7

Out-of-Distribution Semantic Occupancy Prediction

The paper introduces Out-of-Distribution Semantic Occupancy Prediction, a task that focuses on detecting unknown objects in 3D voxel space for autonomous driving. It proposes Realistic Anomaly Augmentation to create two new datasets, VAA-KITTI and VAA-KITTI-360, and presents the OccOoD framework, which uses Cross‑Space Semantic Refinement to improve OoD detection while maintaining semantic occupancy accuracy. Experiments show OccOoD achieves an AuROC of 65.50% and an AuPRCr of 31.83% within a 1.2 m radius, demonstrating strong generalization to real‑world urban scenes.

By Yuheng Zhang, Mengfei Duan, Kunyu Peng, Yuhang Wang, Ruiping Liu, Fei Teng, Kai Luo, Zhiyong Li, Kailun Yang
arXiv Computer Vision
Sep 3

PointGauss: Point Cloud-Guided Multi-Object Segmentation for Gaussian Splatting

PointGauss is a 3D-native framework that performs semantic parsing and instance segmentation on 3D Gaussian splatting representations by treating Gaussian primitives as unstructured point sets and extracting scale‑invariant geometric features with Point Transformer V3. It introduces an adaptive region‑of‑interest cropping strategy and an instance‑aware distance‑constrained rasterization pipeline to enable scalable, view‑consistent pixel‑level projections. The authors also release SplatSeg‑360, a cross‑scale benchmark with 32 complex scenes and over 6,300 aligned 2D‑3D masks, and show that PointGauss achieves real‑time performance with state‑of‑the‑art 3D‑mIoU (~90%) and 2D‑mIoU (~80%) scores.

By Wentao Sun, Yiping Chen, John S. Zelek, Jonathan Li
arXiv Computer Vision
Sep 3

OccAnyScene: Towards Unified Indoor-Outdoor 3D Occupancy Prediction

OccAnyScene introduces a unified approach for 3D occupancy prediction across both indoor and outdoor scenes, addressing the challenge of handling diverse camera setups, spatial ranges, voxel specifications, and semantic taxonomies. The method builds on a pretrained depth model, using pixel-aligned frustum feature aggregation and frustum-parameterized Gaussian construction to generate scene-adaptive occupancy predictions. It achieves state-of-the-art performance, scoring 59.92% mIoU on Occ-ScanNet and 23.06% mIoU on SurroundOcc-nuScenes.

By Junjie Liu, Wanshui Gan, Zitong Dai, Guiping Cao, Yan Li, Ke Chen, Dongmei Jiang, Jianguo Zhang, Xiangyuan Lan
Hugging Face Trending Papers
Sep 8

DXPR: Depth-Based Vision-LiDAR Cross-Modal Place Recognition Using Vision Foundation Models

DXPR is a depth‑based cross‑modal place recognition framework that matches monocular camera queries to a LiDAR map using a single vision foundation model backbone. By converting both modalities into a unified depth image representation, DXPR learns modality‑invariant global descriptors without modality‑specific encoders. A geometry‑aware overlap miner refines pairwise metric learning by computing pixel‑level overlap scores, and extensive tests on KITTI and Boreas show strong performance across seasons, weather, and day/night conditions, outperforming prior CMPR baselines.

Hugging Face Trending Papers
Aug 3

SpatialQuery: Benchmarking Geometry-Grounded Multi-Instance Spatial Reasoning in Vision-Language Models

Vision-language models (VLMs) achieve strong semantic understanding but remain unreliable in metric spatial reasoning, particularly when queries require comparing multiple instances of the same object category. We study this problem through the Closest-Instance Distance Query (CIDQ), where a model must identify the nearest visible candidate to a unique reference object and estimate their gravity-aligned floor-plane distance.