EffectLearner: World-Aware Object-Effect Reasoning for Real-World Video Object Removal

Hugging Face Daily Papers Papers
video-object-removal object-effect-reasoning diffusion-transformer vision-language-model video-editing arxiv

Summary

EffectLearner is a semantic-reasoning-enhanced framework for real-world video object removal, combining a VLM-based Object-Effect Reasoner with a DiT-based Video Eraser and introducing the EffectWorld dataset to handle complex object-induced effects.

Video object removal must eliminate not only the target object but also its induced effects while maintaining high-fidelity and spatiotemporally coherent restoration. Existing methods mainly learn object-effect correspondences implicitly from predefined effect categories and fixed data distributions, limiting their generalization to complex real-world scenes involving compositional effects, spatially detached or weakly correlated effects, long-tail physical phenomena, and dynamically evolving interactions. We propose EffectLearner, a semantic-reasoning-enhanced framework that combines a VLM-based Object-Effect Reasoner with a DiT-based Video Eraser. Guided by a structured effect-analysis prompt, the Reasoner performs cross-modal reasoning over a target-highlighted video and extracts compact effect-aware context, which guides the Video Eraser toward comprehensive object-effect removal. Motion-aware mask guidance and motion-consistency supervision further improve removal coverage and spatiotemporal stability under object motion and evolving scene dynamics. To fully exploit the framework in challenging real-world scenarios, we further construct EffectWorld, a paired video dataset specifically designed for complex object-induced effects, and introduce a progressive training curriculum that combines common supervision with complex-effect data. On the standard ROSE-Bench, EffectLearner outperforms existing baselines on most metrics and achieves clear advantages on both EffectWorld-Eval and the challenging EffectWorld-Wild, demonstrating its ability to deliver high-quality video object removal in complex real-world scenes.
Original Article
View Cached Full Text

Cached at: 08/07/26, 09:56 AM

Paper page - EffectLearner: World-Aware Object-Effect Reasoning for Real-World Video Object Removal

Source: https://huggingface.co/papers/2608.05565 Published on Aug 6

·

Submitted byhttps://huggingface.co/Phoebux

Heon Aug 7

Authors:

,

,

,

,

,

,

,

,

,

Abstract

Videoobjectremovalmusteliminatenotonlythetargetobjectbutalsoitsinducedeffectswhilemaintaininghigh-fidelityandspatiotemporallycoherentrestoration.Existingmethodsmainlylearnobject-effectcorrespondencesimplicitlyfrompredefinedeffectcategoriesandfixeddatadistributions,limitingtheirgeneralizationtocomplexreal-worldscenesinvolvingcompositionaleffects,spatiallydetachedorweaklycorrelatedeffects,long-tailphysicalphenomena,anddynamicallyevolvinginteractions.WeproposeEffectLearner,asemantic-reasoning-enhancedframeworkthatcombinesaVLM-basedObject-EffectReasonerwithaDiT-basedVideoEraser.Guidedbyastructuredeffect-analysisprompt,theReasonerperformscross-modalreasoningoveratarget-highlightedvideoandextractscompacteffect-awarecontext,whichguidestheVideoErasertowardcomprehensiveobject-effectremoval.Motion-awaremaskguidanceandmotion-consistencysupervisionfurtherimproveremovalcoverageandspatiotemporalstabilityunderobjectmotionandevolvingscenedynamics.Tofullyexploittheframeworkinchallengingreal-worldscenarios,wefurtherconstructEffectWorld,apairedvideodatasetspecificallydesignedforcomplexobject-inducedeffects,andintroduceaprogressivetrainingcurriculumthatcombinescommonsupervisionwithcomplex-effectdata.OnthestandardROSE-Bench,EffectLearneroutperformsexistingbaselinesonmostmetricsandachievesclearadvantagesonbothEffectWorld-EvalandthechallengingEffectWorld-Wild,demonstratingitsabilitytodeliverhigh-qualityvideoobjectremovalincomplexreal-worldscenes.

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2608\.05565

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.05565 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.05565 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.05565 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

VideoKR: Towards Knowledge- and Reasoning-Intensive Video Understanding

Hugging Face Daily Papers

VideoKR introduces a large-scale video reasoning dataset and benchmark designed to enhance knowledge-intensive video understanding through expert-domain content and human-in-the-loop example generation. The dataset contains 315K video reasoning examples over 145K expert-domain videos.

LeAct: Learning to Reason from Expert Actions

arXiv cs.CL

LeAct introduces a method to recover chain-of-thought reasoning from silent expert actions by treating the CoT as a latent variable and retaining only those that improve the student's probability of reproducing the action, outperforming baselines in games and robotics.