[WACV 2026] MMHOI: Modeling Complex Multi-Person Multi-Object Interactions

MERL and Mitsubishi Electric researcher Kaen Kogashi presents her paper titled "MMHOI: Modeling Complex Multi-Person Multi-Object Interactions" for the IEEE Winter Conference on Applications of Computer Vision (WACV), to be held in Tucson, Arizona from March 6-10, 2026. The paper was co-authored with MERL researcher Anoop Cherian and Prof. Meng-Yu Jennifer Kuo from the Nara Women’s University, Japan. Paper: https://arxiv.org/pdf/2510.07828 Abstract: Real‑world scenes often feature multiple humans interacting with multiple objects in ways that are causal, goal‑oriented, or cooperative. Yet existing 3D human-object interaction (HOI) benchmarks consider only a fraction of these complex interactions. To close this gap, we present MMHOI -- a large-scale, Multi-human Multi-object Interaction dataset consisting of images from 12 everyday scenarios. MMHOI offers complete 3D shape and pose annotations for every person and object, along with labels for 78 action categories and 14 interaction‑specific body parts, providing a comprehensive testbed for next-generation HOI research. Building on MMHOI, we present MMHOI-Net, an end-to-end transformer-based neural network for jointly estimating human–object 3D geometries, their interactions, and associated actions. A key innovation in our framework is a structured dual-patch representation for modeling objects and their interactions, combined with action recognition to enhance the interaction prediction. Experiments on MMHOI and the recently proposed CORE4D datasets demonstrate that our approach achieves state-of-the-art performance in multi-HOI modeling, excelling in both accuracy and reconstruction quality.