GeoBlur is a framework that estimates the fundamental matrix and relative camera pose from a single motion‑blurred image by exploiting blur artifacts as motion cues. It predicts visual correspondences between two time instances within the exposure window and solves the single‑frame epipolar geometry problem, yielding a fundamental matrix unique up to transposition due to time‑direction ambiguity. The method shows improved performance on synthetic and hybrid benchmarks and remains competitive on real motion‑blur data, also enabling downstream single‑frame motion segmentation.
arXiv:2608.22039v1 Announce Type: new Abstract: Structure-from-Motion (SfM) is a cornerstone of 3D perception, yet current methods often fail when applied to complex videos involving challenging came...
arXiv:2509.04600v2 Announce Type: replace Abstract: Reconstructing global human motion from monocular video is fundamental to VR, graphics, and robotics, yet remains ill-posed due to depth ambiguity,...
arXiv:2603.12064v3 Announce Type: replace Abstract: We address the challenging problem of dense dynamic scene reconstruction and camera pose estimation from multiple freely moving cameras -- a settin...
arXiv:2609.13733v1 Announce Type: new Abstract: Stable and reliable 4D spatial understanding is fundamental for autonomous driving systems. While feedforward reconstruction networks can estimate came...
MINT is a foundation model that directly predicts world-space two-hand trajectories from egocentric RGB video, jointly estimating camera motion, hand states, and hand presence in a single spatiotemporal representation. It uses an open-source labeling pipeline, EGOPIPELINE, to generate large-scale pseudo-labels for pretraining, followed by fine-tuning on a small set of high-quality joint annotations. The model outperforms existing multi-stage approaches in accuracy and speed, and generalizes zero‑shot to unseen egocentric datasets.
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.
Free-viewpoint 3D scene media is increasingly important for immersive applications, yet practical capture often suffers from severe view sparsity and motion blur. Although neural rendering has advanced sparse-view synthesis, existing blur-aware methods typically require substantial multi-view redundancy, accurate camera poses, or costly per-scene optimization.
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.
arXiv:2603.14528v3 Announce Type: replace Abstract: In recent years, 3D visual foundation models, pioneered by pointmap-based approaches such as DUSt3R, have attracted a lot of interest, achieving im...
The paper introduces an affine‑aligned atlas for constructing canonical Gaussian representations in video. By applying frame‑wise affine transforms to absorb global motion before building the canonical Gaussians, the method reduces misalignment between the canonical model and individual frames. This modification can be added to existing Gaussian‑based video representations with minimal extra parameters and improves reconstruction quality, particularly for sequences with large camera motion.
Anchor3R is a streaming 3D reconstruction framework that predicts window-relative poses and local geometry in the current‑frame coordinate system, forming a dense relative‑pose graph for online pose updates and loop‑aware motion averaging. It improves long‑horizon pose accuracy and dense reconstruction quality on indoor, outdoor, driving, and RGB‑D benchmarks, and generalizes from 48‑frame training sequences to streams exceeding 10,000 frames while keeping GPU memory bounded. The method addresses issues of train‑test mismatch, early‑anchor bias, and accumulated drift found in previous streaming models.
arXiv:2609.01276v1 Announce Type: new Abstract: Complete 3D perception from egocentric video requires recovering the surrounding scene and the wearer's full-body motion in a shared metric frame. Exis...