The paper introduces PLANET, a multi‑object tracker that transcends traditional image‑plane limitations by incorporating 3D scene geometry into its query formation. By lifting 2D tracking datasets into 3D and embedding reconstructed geometry into features and positional encodings, PLANET encourages queries to encode object positions. An auxiliary 3D location prediction task and a dual‑resolution temporal memory further enhance performance, enabling state‑of‑the‑art results on three diverse benchmarks.
Monocular videos record 3D scenes as sequences of 2D image-plane projections, obscuring depth and spatial relationships. Multi-object trackers localize and associate objects primarily using appearance...
arXiv:2610.01314v1 Announce Type: new Abstract: Dynamic scenes may be captured by a moving camera, multiple video streams, or images taken at different times. These observations reveal complementary...
TrackEverything is a 3D point tracker that overcomes the trade‑off between sparse long‑horizon tracking and dense short‑clip tracking by representing videos as persistent 3D scene tracks in world coordinates. It introduces voxel‑based de‑duplication at sliding‑window boundaries, a two‑stage refinement process (endpoint refiner and lightweight trajectory refiner), and a 3D WAFT module that replaces memory‑heavy 4D correlation volumes with efficient feature sampling. The method can track all visible points in videos longer than 1000 frames using only 40 GB of GPU memory, outperforming existing dense trackers on short clips and matching sparse trackers on long sequences.
arXiv:2609.18363v1 Announce Type: new Abstract: Online multi camera 3D tracking must maintain scene global identities across synchronized views, yet query-based trackers carry these identities only i...
arXiv:2606. 30576v1 Announce Type: cross Abstract: Cross-view object geo-localization (CVOGL) aims to locate a target object from a query view (e.
Synthesizing a novel-view video from a monocular reference video along a target camera trajectory requires both geometric consistency and motion fidelity with respect to the reference video. Existing methods based on explicit 3D representations are limited by the accuracy of off-the-shelf reconstruction modules, which often produce inaccurate geometry for dynamic objects in monocular videos.
The paper introduces XTrack, a unified framework that repurposes a pretrained video Diffusion Transformer (DiT) for cross-view temporal object correspondence. XTrack incorporates focal cross-view attention alignment, hard negative conditioning, and local context cropping to improve grounding across views and time, and it uses a pseudo‑label curation pipeline to generate dense supervision from sparse Ego‑Exo4D data. The authors demonstrate that XTrack achieves robust temporal consistency and identity preservation in cross‑view tracking tasks.
arXiv:2606. 19733v1 Announce Type: cross Abstract: Efficiently retrieving specific 3D instances from large-scale scenes via natural language prompts remains a formidable challenge in multimedia analysis.
arXiv:2607. 17778v1 Announce Type: cross Abstract: Class-agnostic 3D instance segmentation is critical for robotic systems operating in unknown environments, enabling perception of previously unseen objects for reliable manipulation and navigation.
The paper introduces LingGS, an online monocular RGB Gaussian splatting SLAM system that decouples feed‑forward geometry prediction from camera tracking. By leveraging a frozen streaming 3D foundation model as a virtual RGB‑D sensor for depth estimation and an explicit geometric solver for camera motion via feature reprojection, LingGS achieves higher tracking accuracy than learned predictions. Experiments on indoor and outdoor datasets demonstrate significant improvements in reconstruction quality, achieving about 7 FPS on a single RTX 4090 and 2–3× the throughput of prior SLAM systems.
TQD-Track introduces Temporal Query Denoising (TQD) for 3D Multi‑Object Tracking, extending query denoising beyond single frames by initializing denoising queries from previous‑frame ground truths and propagating them as independent association candidates. The method enriches track queries with temporal context and instance‑specific features, and incorporates diverse noise types to emulate real‑world tracking challenges. Experiments on nuScenes and Argoverse 2 show consistent improvements across multiple MOT baselines with only training‑process modifications.
arXiv:2512. 23020v3 Announce Type: replace-cross Abstract: 3D visual grounding aims to locate objects based on natural language descriptions in 3D scenes.