The recently established 'Out of Sight, Not out of Mind' (OSNOM) task for egocentric videos focuses on tracking objects that are moved by the camera wearer, online, maintaining knowledge of instance locations throughout the video even when they leave the field of view or become heavily occluded. In this paper, we propose the first learning-based solution to the OSNOM task: Whareformer, a transformer-based model with two components: an updatable memory of established tracks and a track assignment module that associates observations with existing tracks in a feed-forward manner.
arXiv:2602. 14771v5 Announce Type: replace-cross Abstract: The human visual system tracks objects by integrating current observations with previously observed information, adapting to target and scene changes, and reasoning about occlusion at fine granularity.
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: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...
arXiv:2610.01201v1 Announce Type: new Abstract: Object permanence, keeping track of an object's identity and position while it is occluded, is central to video representations that track, predict and...
DiDA introduces a lightweight video object segmentation framework that leverages Distillation Learning of Deformable Attention. The method uses deformable attention to adapt key and value positions across frames, enabling object representations that are responsive to spatial and temporal changes. Experiments on DAVIS and YouTube‑VOS benchmarks show state‑of‑the‑art performance and efficient memory usage.
arXiv:2603. 24016v2 Announce Type: replace-cross Abstract: Multi-Object Tracking (MOT) has traditionally focused on a few specific categories, restricting its applicability to real-world scenarios involving diverse objects.
DynaTokens introduces lightweight, scene‑specific tokens that enable camera‑controlled video models to learn and generate dynamic scene motion while keeping the base model frozen. By leveraging cross‑attention, the tokens are trained from a few example trajectories, allowing the model to handle both camera‑induced motion and localized object dynamics. Experiments on VBench2 and WorldScore show that DynaTokens outperforms LoRA, block finetuning, and other trainable‑layer baselines in balancing dynamics and camera control.
arXiv:2603. 09420v3 Announce Type: replace-cross Abstract: Motion forecasting enables autonomous vehicles to anticipate scene evolution by predicting the future trajectories of dynamic agents.
arXiv:2607. 08436v1 Announce Type: cross Abstract: Egocentric human data offers scalable supervision for robot manipulation.
The paper introduces a redundancy-aware fusion framework for EgoExo proficiency estimation, which integrates fine-grained motion cues from egocentric views with spatial context from exocentric views. It identifies multiview redundancy and overfitting as key challenges and proposes two modules—AdaMVS for adaptive view selection and VIB-GB for compressing redundant signals—to address them. Experiments on EgoExo-4D and EgoExo-Fitness show that the method learns to select informative views and fuse them effectively, achieving state‑of‑the‑art results.
arXiv:2609.38347v1 Announce Type: new Abstract: Quantifying collective fish behavior requires accurate trajectories, yet multi-view 3D tracking remains challenging due to frequent occlusions, visuall...
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.