arXiv Computer Vision

FlashReg: GPU-Accelerated 3-Clique Point Cloud Registration for Real-Time Correspondence-to-Pose Estimation

FlashReg is a GPU‑accelerated method for point cloud registration that bypasses the construction of dense second‑order compatibility graphs. It builds a sparse second‑order graph directly from a binary first‑order graph using Fast First‑ and Second‑Order Graph (FFSOG) construction, then performs a dataflow‑optimized 3‑clique search to select correspondence pivots and enumerate triples. On indoor and outdoor benchmarks, FlashReg achieves 2–3× lower correspondence‑to‑pose latency than TurboReg while using roughly half the peak tensor memory on an embedded GPU, making it suitable for high‑throughput onboard perception pipelines.

arXiv Computer Vision
4d ago

Spectral Consistency-Guided Multiview Point Cloud Registration for Low-Overlap Scenes

The paper introduces GMPCR, a non‑learning spectral consistency‑guided framework for multiview point cloud registration in low‑overlap scenes. GMPCR refines initial correspondences into a second‑order compatibility structure, uses spectral analysis to filter unreliable matches and select informative scan pairs, and then applies maximal‑clique hypothesis generation for robust relative transformations. The resulting sparse pose graph is further refined with an adaptive history‑aware synchronization scheme, and a recovery mechanism allows previously down‑weighted edges to regain confidence, achieving high registration recalls on benchmark datasets while reducing computational cost.

By Tianyu Li, Yanghong Lin, Shudong Zhou, Kui Yang, Jingru Zhang, Li Fang, Wei Yao
arXiv Computer Vision
Aug 31

Ex-Sim(3)-Reg: 2D-3D Correspondence Pruning via Extended Sim(3) Registration

The paper introduces Ex‑Sim(3)‑Reg, a fast and robust method for pruning 2D‑3D correspondences by reformulating the problem as an extended Sim(3) registration that explicitly accounts for depth noise. The authors provide a theoretical justification and demonstrate that their approach improves registration recall by up to 24.7% on several benchmark datasets, outperforming state‑of‑the‑art baselines. The code for the method is publicly available on GitHub.

By Pei An, Muyao Peng, Junfeng Ding, Jiaqi Yang, Liangliang Nan
arXiv Computer Vision
Sep 4

Scal3R: Learning Efficient Multi-Relative Pose Query for Scalable Online 3D Reconstruction

Scal3R is a new method for online 3D reconstruction that addresses the failure of traditional models on long videos by decoupling per‑frame depth from global pose estimation. It reformulates reconstruction as a multi‑reference relative pose query, using lightweight learnable tokens (~1% of parameters) injected into a frozen backbone via asymmetric attention to query poses relative to multiple past keyframes. An online pose‑graph optimization with loop closure further suppresses drift, achieving convergence in 8 hours on a single GPU and reducing average absolute trajectory error by over 60% on KITTI while setting state‑of‑the‑art results on several benchmark datasets.

By Chin-Yang Lin, Yang-Che Sun, Cheng Sun, Fu-En Yang, Min-Hung Chen, Yen-Yu Lin, Wei-Chen Chiu, Yu-Lun Liu
arXiv Computer Vision
Aug 26

C3VDReg: A Benchmark for Local-to-Local Colonoscopic Registration toward Anatomical Localization

C3VDReg is a benchmark for local-to-local colonoscopic registration that uses the Colonoscopy 3D Video Dataset (C3VD) to generate 10,015 partial-to-partial point cloud pairs, with 2,088 held‑out test pairs. Each pair consists of a source point cloud from depth reprojection and a target point cloud from CT mesh raycasting, evaluated under a standardized protocol of 8,192 points per cloud and fixed pose conventions. Experiments show that high geometric overlap does not guarantee reliable pose recovery, revealing translation ambiguity along repetitive tubular anatomy as a key failure mode.

By Linzhe Jiang, Jiayuan Huang, Sophia Bano, Matthew J. Clarkson, Zhehua Mao, Mobarak I. Hoque
Hugging Face Trending Papers
Jul 1

Partial Skeleton Visibility for Action Recognition: A Constrained Field-of-View Approach

Skeleton-based action recognition has achieved remarkable success by exploiting joint coordinates and their topological connections, yet prevailing methods overwhelmingly assume complete and clean skeleton inputs. In real-world deployments, such as egocentric vision, crowded surveillance, wearable devices, or edge robotics, limited field-of-view (FoV) frequently causes substantial joint visibility dropout, leading to severe performance degradation that existing models are largely unprepared to handle.