Newest papers

Moving Regions, Not Moving Objects: Segmenting What Actually Moves under Camera Motion

Wenhua Jin, Tiancheng Chen, Sheng Tang, Jiahao Li, Yupeng Liu, Weiqi Zhang, Juntong Fang, Zesong Li · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31209 · cited by 0 works (OpenAlex) · open-access PDF

German

Die bewegliche Objektsegmentierung der beweglichen Kamera wird konventionell auf der Sequenz- und Objektebene formuliert und markiert eine ganze Instanz als bewegend, sobald sie sich irgendwo in einem Video bewegt. Die jüngsten Instant-Formulierungen unterscheiden sich von stationären Intervallen bewegend, generieren aber immer noch Masken auf der Objektebene durch gefrorene Segmentierungs-Grundmodelle. Da die Maske von einer Instanz-Ebene-Vorschlag abgeleitet wird, kann Bewegung in der Regel an die gesamte Instanz verglichen werden, anstatt an die Regionen zu lokalisieren, die sich tatsächlich bewegen. Darüber hinaus können solche Instanz-Ebene-Vorhersagen Masken erzeugen, selbst wenn kein Teil der Instanz sich tatsächlich bewegt, was ein grundlegenderes Problem

Machine translation (Meta M2M100); the English original follows.

Abstract

Moving object segmentation is conventionally formulated at the sequence and object level, marking an entire instance as moving once it moves anywhere in a video. Recent instantaneous formulations distinguish moving from stationary intervals, but still generate masks at the object level through frozen segmentation foundation models. Because the mask is derived from an instance-level proposal, motion is typically attributed to the entire instance rather than localized to the regions that actually move. Moreover, such instance-level predictions can produce masks even when no part of the instance is actually moving, revealing a more fundamental issue: motion is not an intrinsic property of an instance; it can only be determined by comparing at least two frames. Building on this premise, we introduce MoRSe, which jointly interprets two frames to predict a moving-region mask, distinguishing actual scene motion from apparent motion induced by the camera. A single RGB encoder replaces multi-stage foundation stacks, so no optical flow, point tracks, or separate mask generator is needed at inference. We release MoRSe-5K, a benchmark of 42 sequences with Static, Dynamic-Static, and Dynamic settings designed to evaluate the model's ability to distinguish and localize moving and stationary regions under different camera and scene motion conditions. MoRSe outperforms both specialized moving object segmentation methods and dense 4D reconstruction methods with motion-thresholded predictions, while producing the fewest false positives on sequences where nothing moves.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer vision, Artificial intelligence, Segmentation, Computer science, Motion (physics), Object (grammar)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange