arXiv:2609.34302v2 Announce Type: replace Abstract: Multi-camera pedestrian localization is useful for wide-area monitoring in public and commercial spaces. However, deploying these systems often req...
GRACE is a camera‑efficient multi‑view pedestrian tracker that reduces the number of required cameras while maintaining high tracking accuracy. It combines volumetric‑guided fusion of homography‑based BEV features with 3D‑lifted features, uses ray conditioning to incorporate each camera’s viewing direction, and employs BEV Track Recovery to continue existing tracks with low‑confidence detections. On the WildTrack dataset, GRACE raises MOTA from 83.54 to 91.07 compared to the baseline TrackTacular.
Physical AI Smart Spaces is the first benchmark that offers large‑scale, multi‑class, multi‑camera 3D perception data for indoor smart spaces. It includes over 280 hours of synchronized 1080p footage from nearly 1,800 cameras in warehouses, hospitals, and retail venues, with automatic annotations for identities, 2D and 3D bounding boxes, camera calibration, and depth. The benchmark spans synthetic generation, appearance augmentation, and real‑world Sim2Real evaluation, and introduces a 3D version of Higher Order Tracking Accuracy (HOTA) for evaluating multi‑class 3D box tracking.
The paper introduces MovingDroneCrowd++, a large-scale video dataset for dense crowd counting and tracking from moving drones, featuring varied flight altitudes, camera angles, and lighting. It presents two new methods: GD3A for Video Individual Counting and GIA-Track for Multi-Object Tracking, both leveraging group-wise density assignment and identity association to handle aerial challenges. Experiments demonstrate significant improvements, reducing counting error by 47.4% and boosting tracking accuracy by 64.6%.
The paper introduces SAM‑H, a planar object tracker that estimates 8‑degree‑of‑freedom homographies directly from segmentation mask contours using a training‑free pipeline. When applied to masks from SAM 2, SAM‑H achieves a new state‑of‑the‑art performance on the PlanarTrack benchmark, improving the p@5 metric by 18.4 percentage points. The authors also demonstrate that combining segmentation‑based and correspondence‑based homography estimation yields WOFTSAM, which surpasses all previous methods on both PlanarTrack and POT‑210, and provide precise re‑annotations of PlanarTrack initial poses for more accurate benchmarking.
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.
Syn2RealTrack addresses the synthetic‑to‑real gap in multi‑camera 3D perception for warehouses by decomposing it into three distinct issues: camera calibration, object shape prior, and known object census. The pipeline corrects lens distortion from images, fuses detections with a visibility‑weighted part‑based descriptor, measures person height directly from calibration, and uses a closed‑world cardinality prior with a causal filter to eliminate phantom boxes. These local remedies allow the system to adapt without retraining a feature extractor, achieving a 3D HOTA of 52.0118% on the AI City Challenge 2026 Track 1.
The paper introduces a deployment‑focused evaluation protocol for pedestrian tracking that isolates tracker performance by using shared detections, assessing initialization, gap continuation, identity recovery, close‑neighbor association, and runtime under load, while still reporting HOTA as an aggregate metric. Applied to the JackRabbot Dataset, the protocol reveals that trackers vary widely in capability, with none maintaining spatially correct, same‑identity output after 1 s of missing detector support, and that runtime on an NVIDIA Jetson Orin often falls below the 10 Hz target in crowded scenes. The authors provide open‑source code and scripts to enable reproducible assessment of tracker suitability for real‑world pedestrian‑centric robotics. whyItMatters":"The protocol offers a practical, tracker‑only framework that directly informs the design and deployment of mobile robots operating among pedestrians by highlighting specific failure modes and computational constraints that aggregate scores overlook."
VideoReloc presents a method for long‑term indoor video relocalization that relies on a compact semantic scene graph rather than visual appearance. By adaptively selecting clip lengths based on odometry and object‑motion criteria, the system gathers spatial evidence, verifies poses through object triplets, and refines orientation using box faces and gravity cues. This approach achieves high localization accuracy with a tiny 100 kB map, outperforming traditional appearance‑based methods on RIO10 and ReplicaCAD datasets.
Deep-learning features excel in visual matching, yet their practical value in tightly coupled visual-inertial SLAM (VI-SLAM) remains insufficiently characterized. We present DL-VINS-Factory, a unified framework that integrates learned feature extractors (ALIKED, RaCo, SuperPoint, XFeat) with either Lucas--Kanade (LK) optical-flow tracking or LightGlue (LG) descriptor matching.
UniTrackPLA introduces a unified panorama-language-action model that simultaneously handles instruction‑guided navigation and dynamic person tracking for embodied robots. Its Panoramic‑Aware Encoding preserves azimuthal and temporal structure, allowing a shared vision‑language backbone to generate continuous waypoint chunks for both tasks. The model also employs World‑Action Consistency to predict future visual states and verify waypoint prefixes, enabling reliable action reuse and replanning when inconsistencies arise. A new OmniTrackNav‑Bench dataset and extensive real‑world experiments demonstrate significant performance gains over prior methods.
arXiv:2509. 08421v2 Announce Type: replace-cross Abstract: For multimedia spatial intelligence through time, multi-view multi-object tracking (MVMOT) suffers from persistent challenges in maintaining consistent object identities across different camera views, leading to tracking inaccuracies.