AI TrackerFrontier desk · AEST
Back to news
normalaitech

MAMHOI: Factorizing Scene-Aware Human-Object Interaction through Affordances

arXiv cs.AI · 9 Oct 2026, 03:52 AEST

MAMHOI introduces a method for generating realistic human-object interactions in complex 3D scenes. It utilizes a factorization approach that enhances scene understanding and motion synthesis without needing paired human-object-scene data.

Key facts

Read aloud

MAMHOI introduces a method for generating realistic human-object interactions in complex 3D scenes. It utilizes a factorization approach that enhances scene understanding and motion synthesis without needing paired human-object-scene data. Generating realistic human-object interactions HOI in complex 3D scenes requires two complementary capabilities: reasoning about interaction feasibility in the environment and synthesizing realistic human-object motion. Human-scene datasets provide rich information about environment-aware motion, while human-object datasets capture detailed interaction dynamics, yet paired human-object-scene data remain scarce. MAMHOI factorizes scene-aware HOI generation through an explicit motion-affordance interface between scene understanding and motion synthesis. Experiments in complex indoor environments show that MAMHOI reduces object--scene penetration while better preserving human--object interaction quality, yielding more realistic and physically feasible scene-aware interactions.

Sources in this cluster

MAMHOI: Factorizing Scene-Aware Human-Object Interaction through Affordances · AI Tracker