MIRAGE is a method that fuses feed‑forward visual geometry with LiDAR‑IMU metric motion to provide reliable metric pose estimation and high‑fidelity Gaussian reconstruction. It regulates pose corrections based on correspondence support and cross‑modal agreement, achieving accurate metric trajectories and outperforming COLMAP on datasets such as Oxford Spires, M2DGR, and INS. Additionally, MIRAGE was used to create a new multimodal data‑collection platform and establish accurate 6‑DoF reference trajectories for a vision‑only benchmark, enabling systematic evaluation of monocular large‑scale reconstruction methods in long‑horizon indoor scenes.
By Xinzhe Wang, Changjian Jiang, Kaiwen Song, Xudong Li, Kerui Ren, Ran Yi, Lizhuang Ma, Chunhua Shen, Linning Xu, Tao Lu, Mulin Yu
The paper introduces CalfVO, a monocular visual odometry system that operates without camera intrinsics, test‑time optimization, bundle adjustment, or loop closure. Using a transformer, it predicts relative poses with separate rotation and translation confidences over overlapping image windows, then aggregates these predictions via a confidence‑weighted module to produce a single trajectory. CalfVO achieves the highest accuracy among calibration‑free methods across five benchmarks and runs at 53 FPS, outperforming all baselines.
By Vladimir Yugay, Duy-Kien Nguyen, Theo Gevers, Cees G. M. Snoek, Martin R. Oswald
arXiv:2506. 22784v2 Announce Type: replace-cross Abstract: Point-pixel registration between LiDAR point clouds and camera images is a fundamental yet challenging task in autonomous driving and robotic perception.
By Yu Han, Zhiwei Huang, Yanting Zhang, Fangjun Ding, Shen Cai, Xiaoyu Tang, Yanchao Dong, Rui Fan
DefVINS is a visual‑inertial odometry pipeline tailored for deformable scenes, breaking the rigidity assumption of traditional VIO. It decomposes the odometry state into a rigid, IMU‑anchored component and a non‑rigid scene warp using an embedded deformation graph. The authors also introduce VIMandala, the first real‑world benchmark with ground‑truth camera poses for deformable VIO, and extend the synthetic Drunkard’s benchmark with inertial data, demonstrating that DefVINS outperforms both rigid and non‑rigid baselines.
By Samuel Cerezo, Javier Civera
M3GD introduces a multimodal representation that fuses pre‑trained 2D image and 3D LiDAR foundation models for robotic novel view synthesis, avoiding the need for a separate cross‑modal translator. By projecting LiDAR onto the image latent grid and injecting the resulting geometry‑aware packets via a lightweight residual adapter, the method enhances both RGB and depth synthesis on the GrandTour dataset compared to an image‑only baseline. Ablation studies confirm that pixel‑aligned LiDAR content drives the performance gains, and real‑world deployment on a ground robot demonstrates a tunable quality–cost trade‑off.
By Yang Zhou, Jiuhong Xiao, Shizhao Ye, Long Quang, Carlos Nieto-Granda, Giuseppe Loianno
The paper introduces TempLoc, a Temporal‑aware Localization framework that improves outdoor LiDAR relocalization by leveraging spatio‑temporal consistency across scans. It first predicts point‑wise global coordinates with uncertainties, then estimates inter‑frame correspondences using an attention‑based Prior Coordinate Generation module, and finally fuses these predictions in an uncertainty‑guided manner to produce a more accurate global 6‑DoF pose. Experiments on the NCLT and Oxford RobotCar datasets show that TempLoc significantly outperforms existing state‑of‑the‑art methods.
By Minghang Zhu, Zhijing Wang, Yuxin Guo, Chen Liu, Yongshu Huang, Wen Li, Sheng Ao, Cheng Wang
DAVIO is a dense monocular‑inertial SLAM system that leverages a single multi‑view depth model (Depth Anything 3) for both initialization and mapping. It starts up quickly by solving a feature‑free linear system from a five‑image window and IMU pre‑integration, then uses a VIO filter whose metric poses condition the depth model during tracking. The system corrects residual scale along viewing rays, preserves metric baselines, and refines the map with a gravity‑preserving sub‑map graph, achieving earlier start‑up, lower localization error, and more accurate dense maps than state‑of‑the‑art feed‑forward mappers on both EuRoC and building‑scale ORI datasets.
By Jaafar Mahmoud, Arthur Movsesyan, Mikhail Iumanov, Sergey Kolyubin
TAPVid-MV is a new benchmark for tracking any point in 3D across multiple synchronized camera views. It comprises 284 sequences, 1,142 calibrated camera streams, and 109,769 point tracks, covering indoor and outdoor domains and derived from various modalities such as depth, LiDAR, SLAM, and simulation. The dataset is visually verified, and evaluation shows that current multi‑view trackers do not consistently outperform monocular trackers, highlighting geometry recovery as a key bottleneck.
By Skanda Koppula, Frano Rajic, Abdullah Faiz Ur Rahman, Yi Yang, Ignacio Rocco, Jeet Thakwani, Rishabh Kabra, Andrew Zisserman, Joao Carreira, Siyu Tang, Carl Doersch, Gabriel Brostow
arXiv:2610.28571v1 Announce Type: new
Abstract: ScreenCalib enables light detection and ranging (LiDAR)--camera extrinsic calibration using an ordinary programmable display, providing known visual ma...
By Zuhao Zou, Xiangcheng Hu, Jianhao Jiao, Jin Wu
M3GD is a novel approach for robotic novel view synthesis that fuses camera images and LiDAR point clouds without requiring a separate cross‑modal translator. By projecting LiDAR data onto the image latent grid and injecting it via a lightweight residual adapter, M3GD enhances both RGB and depth generation on the GrandTour dataset compared to image‑only baselines. Experiments on a ground robot confirm that the method can be deployed in real‑world scenarios with a tunable quality‑cost trade‑off.
arXiv:2609.22716v1 Announce Type: new
Abstract: Image-to-LiDAR registration estimates the camera pose of an image with respect to a LiDAR point cloud. It has diverse applications in autonomous drivin...
By Zijun Li, Xiaotian Sun, Xuelun Shen, Yao Dai, Sheng Ao, Yangyang Shi, Jakob Engel, Zhipeng Cai, Cheng Wang
DRS‑VPT is a feed‑forward transformer that performs image‑to‑scan registration by predicting the scan pose, point maps, and a coarse‑to‑fine feature pyramid for direct reprojective alignment. It unifies tasks like camera‑LiDAR calibration and indoor camera‑to‑map relocalization, achieving state‑of‑the‑art results in autonomous driving and competitive performance indoors without map‑specific training. The model also learns complex scan‑to‑image projection behaviors, such as occlusion of back‑facing points.
By Lanke Frank Tarimo Fu, Maurice Fallon