Skip to content
AI News HubLIVE
Original source2 min read

What Moves? Localized Motion Representations for Compositional Scene Control

Summary

The paper introduces a promptable localized motion representation that produces persistent, region-addressable embeddings for user-specified video regions while retaining global context, enabling object-level motion transfer and localized action classification, and outperforming cropping- or masking-based localization.

SourcearXiv Computer VisionAuthor: Frank Fundel, Malek Ben Alaya, Thomas Ressler-Antal, Stefan Andreas Baumann, Bj\"orn Ommer
What Moves? Localized Motion Representations for Compositional Scene Control
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 3 Sep 2026]

Title:What Moves? Localized Motion Representations for Compositional Scene Control

View a PDF of the paper titled What Moves? Localized Motion Representations for Compositional Scene Control, by Frank Fundel and 4 other authors

View PDF HTML (experimental)

Abstract:Real-world dynamics are inherently compositional: multiple entities move simultaneously within a shared scene, each exhibiting distinct motion patterns. Yet most existing video representations encode motion globally, without explicitly capturing localized motion for individual entities. Crucially, motion is defined relative to a global reference frame, including camera motion and scene layout. However, localized embeddings are often computed from cropped images or obtained by masking features after encoding, discarding the context needed to interpret motion. To address this, we introduce a promptable localized motion representation that produces persistent embeddings for user-specified regions defined by spatial masks. Rather than cropping the input or masking features, our model processes the full video and conditions motion encoding directly on the queried region. This yields temporally consistent, region-addressable embeddings that isolate local dynamics while retaining the global context required for disambiguation. We demonstrate object-level motion transfer, enabling controlled composition of dynamic scenes. Beyond generative control, our embeddings support localized action classification in multi-actor videos. Across both tasks, our approach improves controllability and outperforms global representations localized through cropping or post-hoc masking. Project Page: this https URL

Subjects:

Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)

Cite as: arXiv:2609.04383 [cs.CV]

(or arXiv:2609.04383v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2609.04383

arXiv-issued DOI via DataCite (pending registration)

Journal reference: Proceedings of the European Conference on Computer Vision (ECCV), 2026

Submission history

From: Frank Fundel [view email] [v1] Thu, 3 Sep 2026 18:44:53 UTC (37,108 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled What Moves? Localized Motion Representations for Compositional Scene Control, by Frank Fundel and 4 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-09

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • A promptable localized motion representation conditions motion encoding on spatial masks over the full video, avoiding cropping or post-hoc feature masking.
  • The resulting embeddings are temporally consistent and region-addressable, isolating local dynamics while retaining camera motion and scene layout context.
  • The method supports object-level motion transfer for controlled dynamic scene composition and localized action classification in multi-actor videos.
  • The work is published at ECCV 2026 and authored by Frank Fundel and four co-authors.

Highlights and analysis are generated automatically and may contain errors. Check the original source.